2007 NORTHEASTERN NATURALIST 14(3):425–438 Arboreal Mite Communities on Epiphytic Lichens of the Adirondack Mountains of New York Heather T. Root1,2,*, Gregory G. McGee1, and Roy A. Norton1 Abstract - We describe the mite fauna inhabiting the canopies of remnant oldgrowth Acer saccharum (sugar maple) trees in northern hardwood stands under different silvicultural treatments in the Adirondack Mountains. We also compare mites on different arboreal substrates, including bare bark, the crustose lichen Pertusaria velata, and three foliose lichens: Flavoparmelia caperata, Parmelia squarrosa, and Punctelia rudecta. A total of 877 individual mites were collected representing 25 oribatid mite species, at least three of which are undescribed, and nine non-oribatid mite families. Mite abundance was sevenfold greater in Punctelia rudecta than on bare bark, and communities differed among bark, crustose lichen, and foliose lichens, but not among the different species of foliose lichens. Trees in old growth and reserve shelterwood stands supported different mite communities. Introduction Mites are ubiquitous in forested ecosystems and contribute substantial biodiversity to these habitats (Behan-Pelletier and Walter 2000, Behan- Pelletier and Winchester 1998, Seyd and Seaward 1984, Winchester et al. 1999). In soil ecosystems, they have been used as sensitive environmental indicators (André et al. 1982, Behan-Pelletier 1999, van Straalen 1998) and may also be indicative of disturbances related to forestry practices in terrestrial ecosystems (Winchester et al. 1999). In addition, mites associated with arboreal lichens and mosses in the Pacific Northwest contribute to the canopy ecosystem by altering dynamics of nitrogen and other nutrients (Carroll 1980) and play important roles in canopy food webs (for review see Pettersson et al. 1995, Walter and Proctor 1999). Canopy mite communities of northern temperate forest systems have been well-studied in Europe (André 1985, Nicolai 1986, Prinzing 2005, Seyd and Seaward 1984, Travé 1963, Wunderle 1992), where some species are associated with particular lichen habitats (Fröberg et al. 2003, Seyd and Seaward 1984) or tree species (Nicolai 1986). Mite communities differ among arboreal microhabitats both spatially and temporally in Belgian and German forests (André 1985, Wunderle 1992). North American studies are fewer, but recent canopy work in the Pacific Northwest has also uncovered great oribatid mite diversity (Carroll 1980, Lindo and Winchester 2006, 1State University of New York College of Environmental Science and Forestry, Faculty of Environmental and Forest Biology, 1 Forestry Drive, Syracuse, NY 13210. 2Current address - Oregon State University, Department of Botany and Plant Pathology, Corvallis, OR, 97331-2902. *Corresponding author - rooth@science.oregonstate.edu. 426 Northeastern Naturalist Vol. 14, No. 3 Winchester et al. 1999). Arboreal oribatid mites have not been investigated in northeast North America, except in studies of particular taxa, such as Carabodes (e.g., Reeves 1988), or where arthropods were identified in broad taxonomic categories (Pettersson et al. 1995, Stubbs 1989). Lichen-dwelling mites commonly are specific to habitats and require lichen cover for shelter and food (Seyd and Seaward 1984). Because epiphytic cover is related to abundance, diversity, and composition of microarthropod communities (André 1985, Fröberg et al. 2003, Nicolai 1986), and because epiphytes can be affected by a variety of disturbances (e.g., Cameron 2002, McCune 1993, Richardson and Cameron 2004), knowledge of how resident arthropod communities are related to epiphyte assemblages may be important in understanding ways to conserve both taxa in northeast North America. The objectives of our research were twofold: (1) to provide the first description of lichen-dwelling arboreal mite communities on remnant old-growth Acer saccharum Marshall (sugar maple) in the Adirondack Mountains, with a focus on oribatid mites; and (2) to investigate the habitat associations of arboreal mite species in crowns of sugar maple trees. To address the second objective, we tested two hypotheses: (1) total mite abundance and community composition differs among five arboreal substrates: bare bark, the crustose lichen Pertusaria velata (Turner) Nyl., and three foliose lichens: Flavoparmelia caperata (L.) Hale, Parmelia squarrosa Hale, and Punctelia rudecta (Ach.) Krog., and (2) overall mite abundance and community composition differs between large trees in reserve shelterwood stands and those in old growth. Methods Research was conducted at the State University of New York College of Environmental Science and Forestry’s Huntington Wildlife Forest (HWF) in Newcomb, Essex County, NY, USA. Based on unpublished data collected from 1940 to 2004, mean annual precipitation at HWF is 104 cm, snowfall is 295 cm and mean monthly temperatures range from -9.4 ºC in Jan. to 18.3 ºC in July, with a growing season of 120 days. The HWF is set within the Adirondack hemlock-white pine-northern hardwood forest type described by Braun (1950). These systems are dominated by sugar maple, Fagus grandifolia Ehrh. (American beech), Betula alleghaniensis Britton (yellow birch), and the conifers Tsuga canadensis (L.) Carrière (eastern hemlock), Pinus strobus L. (eastern white pine), Abies balsamea L. Miller (balsam fir) and Picea rubens Sarg. (red spruce). Less frequent hardwood species include Prunus serotina Ehrh. (black cherry), Tilia americana L. (basswood), Fraxinus americana L. (white ash) and Acer rubrum L. (red maple). Mites were sampled from June 22 to August 7, 2005, in three old-growth and three reserve shelterwood stands (Nyland 2002). The old-growth stands had no recorded fire or management history except minimal red spruce removal in the mid-1800s. Their UTM NAD 83 Zone 18 coordinates are 2007 H.T. Root, G.G. McGee, and R.A. Norton 427 (563061, 4868590), (560651, 4870410), and (558513, 4872822). They contained large upper-canopy trees of up to 90-cm dbh and had basal areas ranging from 22 to 32 m2 ha-1. Despite the lack of harvesting activity, these stands are located within the aftermath zone of the beech bark disease complex and therefore their structures have been substantially altered (Forrester et al. 2003, Shigo 1972). Reserve shelterwoods were former oldgrowth stands regenerated using even-aged shelterwood methods during the 1970s. Their UTM NAD 83 Zone 18 coordinates are (560418, 4870364), (560553,4875259), and (565856, 4871410). Since seed-trees were not removed, these residual trees form a dispersed and discontinuous overstory under which the new, even-aged cohort developed (Nyland 2002). Residual basal areas following shelterwood cuts ranged from 6 to 14 m2 ha-1. Two large (>55 cm dbh) sugar maple trees were randomly selected in each stand for mite sampling. Sampling only one tree species allowed for constancy of bark characteristics, which may affect mite communities by offering differing microclimates in the texture and color of bark (Nicolai 1986). Trees were climbed using standard arborist techniques and samples were taken from large (>10 cm diameter) branches in the lower to middle parts of the crown. Samples were taken from bare bark, Pertusaria velata, Parmelia squarrosa, Punctelia rudecta, and Flavoparmelia caperata. These species were common and consistently present on all large trees. Samples were taken by hammering a 5.3 cm diameter sharpened steel cylinder, with a striking plate welded to the unsharpened end, through the lichen and bark to the cambium. Each sample consisted of two cores with a combined total area of 44 cm2. Care was taken to place the cylinder over a large patch of lichen or bare bark such that only one substrate was sampled. Each of the five substrates was sampled on two trees in six stands for a total of 60 samples. In the laboratory, samples were emptied into small modified Tullgren funnels and extracted into 70% ethanol for approximately 24 hours. The funnels were 9-cm tall x 10-cm diameter plastic plant pots fitted with window screens at the base. A 7-watt light bulb was positioned 1 cm from the top of each funnel. After Tullgren extraction, lichen and bark samples were transferred to a weak KOH solution (c. 0.17 g KOH per 1 L water) to soak for 24 hr. Soaked lichens and bark were washed with water, from which mites were collected using a 75-􀀫m sieve. Mites from each extraction method were then combined, sorted from fine sample debris under a dissecting microscope, cleared in lactic acid and studied in cavity slides with both bright field and differential interference contrast illumination. Non-oribatid mites were identified to family using Dindal (1990) or Krantz (1978). Oribatid mites were identified to genus with an unpublished key by V. Behan-Pelletier and R.A. Norton. Species were identified by comparison to known specimens, original descriptions, monographs, and other primary taxonomic literature. Nomenclature follows Marshall et al. (1987) except for Carabodes, which follows Reeves and Behan-Pelletier (1998). 428 Northeastern Naturalist Vol. 14, No. 3 Differences in total mite densities among substrates and stand types were tested along with sample date using a linear regression, with substrate and stand type as class variables and number of days since June 22 as a continuous variable (proc glm; SAS Institute [2005]). Density data were log-transformed to meet assumptions of normality and equal variance. Lichen community composition differences among substrates were tested using PC-ORD (McCune and Mefford 1999), a program designed for analyzing ecological communities and their relationships with environmental variables. First, a species accumulation curve was produced to assess adequacy of sampling. Then, a non-metric multidimensional scaling (NMS) ordination (Kruskal 1964) using Sørenson’s distance (Bray and Curtis 1957) on the raw count data with slow and thorough autopilot mode was performed to visualize and interpret multivariate data. Ranked MRPP was used to determine whether mite communities differed among stand types and substrates. Ranked MRPP is a nonparametric procedure that tests differences among ranked distances to differentiate among pre-existing groups. This procedure produces a P value as well as an A value. The latter can be interpreted as an effect size: when A is close to zero, groups are no more different than expected by chance. Higher A values indicate progressively more meaningful differences between groups. In community ecology, A values can be less than 0.1 even when groups are clearly ecologically different and an A > 0.33 is exceptionally high (McCune and Grace 2002). Pairwise comparisons among the five substrates were also made using this method. Indicator species analyses using Monte Carlo tests with 1000 runs were also performed in PC-ORD to determine which species were significantly associated with particular substrates or stand types. These analyses produced P-values and indicator values. An indicator of 100 should be interpreted to mean that the species is always present in one particular habitat and never present in the others (McCune and Grace 2002). Lichen community composition differences between stand types were tested using total mite counts in each of the stands; therefore, instead of 60 sample units, these were combined into the six stands sampled. As with the substrate tests, a ranked MRPP determined whether stand types differed significantly, followed by an indicator species analysis to determine which mite species were associated with the stand types. Results A total of 25 oribatid mite species and nine families outside of Oribatida were represented in the samples (Table 1). We collected 877 individuals in Table 1 (facing page). Frequency, mean, and standard deviations of mite density (per 44 cm2), and total number of individuals sampled for respective mite taxa collected from Acer saccharum (sugar maple) canopies in Adirondack northern hardwood old-growth and reserve shelterwood stands (n = 60 samples). Oribatid mites are adults unless otherwise noted; immatures were not determined to species. # = number collected. 2007 H.T. Root, G.G. McGee, and R.A. Norton 429 Frequency in 6 in 60 Mean Taxon stands samples density St. # Prostigmata Bdellidae 3 5 0.09 0.29 5 Cunaxidae 2 2 0.04 0.19 2 Erythraeidae 2 2 0.04 0.19 2 Tydeidae 1 1 0.02 0.13 1 Mesostigmata Ascidae 4 6 0.11 0.31 6 Ologamisidae 3 5 0.18 0.66 10 Laelapidae 2 3 0.05 0.23 3 Astigmata Acaridae 1 1 0.14 1.07 8 Glycyphagidae 2 2 0.04 0.19 2 Oribatida Achipteriidae immature 5 21 1.20 2.28 67 Adrodamaeus musci (Paschoal) 1 1 0.04 0.27 2 Anachipteria magnilamellata (Ewing) 3 7 0.71 2.16 40 A. n. sp.A 5 35 4.60 7.53 257 Caleremaeus sp. 1B 4 8 0.20 0.52 11 Camisia horrida (Hermann) 3 7 0.20 0.72 11 C. sp. immature 3 5 0.20 0.72 11 Carabodes brevis (Banks) 1 1 0.07 0.54 4 C. higginsi (Reeves) 6 11 2.43 4.08 136 C. radiatus (Berlese) 3 14 0.43 0.93 24 C. spp. immature 1 1 0.02 0.13 1 Ceratoppia bipilis (Hermann) 1 4 0.13 0.47 7 C. sp. immature 1 1 0.02 0.13 1 Cultrobates quadricuspidatus (Ewing) 4 4 0.07 0.26 4 Epidamaeus sp. 1C 4 5 0.13 0.47 7 E. sp. 2 1 1 0.02 0.13 1 Eueremaeus columbianus (Berlese) 3 5 0.10 0.29 5 E. marshalli (Behan-Pelletier) 1 2 0.04 0.19 2 E. sp. immature 1 1 0.02 0.13 1 Eupelops latipilosus (Ewing) 6 9 0.18 0.43 10 E. sp. immature 1 1 0.02 0.13 1 Hemileius quadripilis (Fitch) 6 23 0.66 0.96 37 H. sp. immature 2 2 0.04 0.19 2 Ommatocepheus clavatus (Wooley & Higgins) 4 8 0.32 1.49 18 O. sp. immature 4 7 0.34 1.33 19 Oribatella sp. 1 1 2 0.04 0.19 2 Phauloppia banksi (Marshall, Reeves & Norton) 6 35 1.63 1.99 91 Phthiracarus bryobius (Jacot) 3 4 0.11 0.41 6 Platyliodes sp. 1D 2 4 0.07 0.26 4 P. sp. immature 3 8 0.21 0.59 12 Protokalumma depressa (Banks) 1 1 0.02 0.13 1 Scapheremaeus palustris (Sellnick) 4 9 0.27 0.80 15 S. sp. immature 3 7 0.14 0.40 8 Scheloribates spp. 2 4 0.11 0.41 6 Zygoribatula exilis (Nicolet) 5 12 0.25 0.51 14 AAnachipteria n. sp. is a new species with saccules similar to A. dubia (Weigmann) except with globose sensilli. BCaleremaeus sp. 1 is a new species differing from both C. monilipes and C. retractus. CEpidamaeus sp. 1 is a new species of Epidamaeus with no spinae adnate that retains its scalps. E. sp. 2 was only encountered once and was not determined to species. DPlatyliodes sp. 1 is probably a new species. It is most similar to P. macroprionus (Wooley et Higgins) decribed from Washington. 430 Northeastern Naturalist Vol. 14, No. 3 total, with an average of 14.6 and a maximum of 54 mites per 44 cm2 sample. Only Carabodes higginsi, Hemileius quadripilis, and Eupelops latipilosus were found in all six stands. Anachipteria n. sp., Carabodes higginsi, and Phauloppia banksi were the most numerous species with total counts of 257, 136, and 91, respectively. The taxon-accumulation curve had a final slope of 13%, and the first-order jackknife estimate of species richness was 54 taxa. Anachipteria n. sp. was most numerous and was collected on 10 of 12 sampled trees and in five of the six stands (taxonomic description of this species is in preparation). Total mite abundance differences among substrate types and stand types were tested simultaneously. Stand type and all related interactions were not significantly related to mite density and were therefore omitted from the analysis. Total mite density declined slightly, but significantly (P = 0.05), as daily sampling progressed into the summer after June 22 (Table 2). Mite densities differed among the sampled substrates (P < 0.001). Bare bark supported the lowest observed mite density, while the foliose lichen Punctelia rudecta supported mite density that was seven-fold greater (Table 3). Mean mite densities did not differ between crustose lichen and bare bark. Although Flavoparmelia caperata had only 59% of the mean mite density of Punctelia rudecta, these foliose lichens were so variable in their total mite counts as to be indistinguishable. Other research by the senior author (Root et al., in press) indicates that the estimated areas covered by the three foliose lichens on sugar maple Table 3. Means and standard deviations of total mite densities (number of individuals per sample) on five substrates sampled in Acer saccharum (sugar maple) canopies in Adirondack northern hardwood old-growth and reserve shelterwood stands (n = 60). Groupings marked with the same letter do not differ significantly (alpha < 0.05), correcting for experiment-wise experimental error using Bonferroni’s method. Substrate Mean St. Grouping Bark 4.1 6.1 A Pertusaria velata 4.3 3.0 A Flavoparmelia caperata 17.1 15.5 B Parmelia squarrosa 18.8 12.6 B Punctelia rudecta 28.8 21.3 B Table 2. General linear model on log-transformed mite densities on Acer saccharum (sugar maple) in Adirondack northern hardwood old growth and reserve shelterwood stands. Factors considered are day since June 22 and substrates (including bark, Pertusaria velata, Flavoparmelia caperata, Parmelia squarrosa, and Punctelia rudecta). Source DF SS MS F P Model 9 41.7 4.6 7.1 less than 0.001 Day 1 2.6 2.6 4.0 0.05 Substrate 4 34.3 8.6 13.1 less than 0.001 Day*Substrate 4 4.8 1.2 1.8 0.14 Error 50 32.7 0.65 Total 59 74.4 2007 H.T. Root, G.G. McGee, and R.A. Norton 431 trunks and branches greater than 10 cm diameter in these stands were 32.4, 73.8, and 69.2 m2 ha-1 for Flavoparmelia caperata, Parmelia squarrosa, and Punctelia rudecta, respectively. Multiplying these areas by the average mite densities extrapolates this estimate to 125,000 on F. caperata, 315,000 on P. squarrosa, and 452,000 mites per hectare on P. rudecta. Confidence in these estimates is low, since variability is large in both the mite and lichen cover estimates. The NMS ordination of all sample units (Fig. 1) showed a distinct cluster of foliose lichen samples toward the center, with crustose lichens and bark on the periphery. This ordination should be interpreted with some caution because of its high stress and the low variation accounted for (final stress = 27.0, instability = 0.0001 with 400 iterations, overall r2 = 0.57, axis-1 r2 = 0.29, and axis-2 r2 = 0.28). There was no apparent relationship between sample day and mite community composition. However, individual species abundances seemed related to date; for example, Anachipteria n. sp. decreased over the summer whereas Carabodes radiatus increased. Figure 1. NMS ordination of sample units as obtained from mite count data for 44 mite taxa from various substrates on Acer saccharum (sugar maple) in Adirondack northern hardwood old-growth and reserve shelterwood stands. Crustose lichen samples are represented by crosses, whereas samples from the three foliose substrates are marked with filled icons. 432 Northeastern Naturalist Vol. 14, No. 3 The overall ranked MRPP distinguished mite communities among the five substrates (P < 0.0001, A = 0.13). Pairwise comparisons (Table 4) indicated that mite community composition on bark differed from crustose and all foliose lichens. Mite community composition on crustose lichen was discernible from that of bark and all foliose species. However, there were no differences in the mite communities among the three different foliose lichen species. Indicator species analysis among bark, crustose, and foliose lichens yielded four significant indicator species for the various substrates (Table 5). Mite community composition, unlike total density, differed between reserve shelterwoods and old growth (n = 3 stands per type, A = 0.206, P = 0.023) and five taxa emerged as indicative of a particular stand type (Table 5). Discussion A total count of 25 Oribatida species is similar to estimates given by other researchers who worked at similar latitudes and sampled epiphytes and Table 5. Indicator-species analysis on mite count numbers on five substrates of 44 taxa on 56 non-empty samples Acer saccharum (sugar maple) in Adirondack northern hardwood oldgrowth and reserve shelterwood stands. Second portion of the table represents indicator species between two stand types (n = 3 stands/stand type): old growth and reserve shelterwood. Overall experimental error is not controlled. Indicator Mite species Substrate or stand type value P value Achipteriidae imm. Foliose lichen 47.6 0.004 Anachipteria n. sp. Foliose lichen 59.1 0.002 Carabodes higginsi Foliose lichen 70.5 0.001 Ommatocepheus sp. imm. Bark 31.2 0.013 Achipteriidae imm. Reserve shelterwood 70.1 0.102 Anachipteria magnilamellata Old growth 100.0 0.102 A. n. sp. Reserve shelterwood 75.1 0.102 Caleremaeus sp. 1 Reserve shelterwood 81.8 0.102 Platyliodes sp. imm. Old growth 100.0 0.102 Table 4. Ranked MRPP statistics on mite count numbers of 44 taxa on 56 non-empty samples from Acer saccharum (sugar maple) in Adirondack northern hardwood old-growth and reserve shelterwood stands. Overall experimental error is not controlled. Comparison A value P value Bark: Pertusaria velata 0.06 0.03 Bark: Flavoparmelia caperata 0.12 <0.001 Bark: Parmelia squarrosa 0.14 <0.001 Bark: Punctelia rudecta 0.18 <0.001 Pertusaria velata: Flavoparmelia caperata 0.06 0.02 P. velata: Parmelia squarrosa 0.09 0.003 P. velata: Punctelia rudecta 0.17 <0.001 Flavoparmelia caperata: Parmelia squarrosa -0.01 0.54 F. caperata: Punctelia rudecta 0.02 0.20 Parmelia squarrosa: Punctelia rudecta -0.004 0.47 2007 H.T. Root, G.G. McGee, and R.A. Norton 433 bark in hardwood forests. On tree bases of hardwoods in Minnesota, 19 oribatid mite species were collected (Nicolai 1993). Collections of nearly 19,000 oribatid mites from hardwood trees in Belgium represented 36 species (André 1984), whereas in Poland, 35 species were collected in a Pinus sylvestris L. (Scots pine) forest near a copper smelting plant (Seniczak et al. 1996). In a British Columbian Picea sitchensis (Bong.) Carr. (sitka spruce) forest, 36 Oribatida species were collected (Behan-Pelletier and Winchester 1998). In Germany, intensive sampling with a variety of collecting methods in a Fagus sylvatica L. (European beech) forest yielded 64 oribatid species on a variety of arboreal substrates (Wunderle 1992). The species accumulation curve and jackknife estimates indicate that sampling was not sufficient to describe species richness precisely. We estimated the sample number of 44-cm2 samples required to achieve a precision of standard error within 15% of the mean for Anachipteria n. sp., Carabodes higginsi, and Phauloppia banksi to be 119, 125, and 66 respectively. Future arboreal mite sampling aimed at precise species diversity and density estimation in this region should double our sampling intensity. Total mite counts declined with date after June 22, but this pattern was variable among species, perhaps because of temporal patterns in the life cycle of the mites, usage of seasonal resources, or responses to weather events. Future studies aimed solely at obtaining the maximum number of mites might focus on early summer, but those attempting to describe the full arboreal mite community should have more intense sampling spread through the entire summer or even the whole year (for example, André 1985). Further work is necessary in northeast North America to determine arboreal oribatid mite life-cycle timing and resource usage. In this preliminary study, we sampled a total of only 0.26 m2 surface area on 12 trees and found at least three undescribed oribatid mite species. New species and new records are not uncommon when sampling arboreal habitats (Behan-Pelletier and Walter 2000) because canopy ecosystems remain largely unexplored. Further canopy mite studies in this region should yield more information about this wealth of mite diversity associated with epiphytes of northern hardwood forests. Our finding similar mite communities on different foliose lichen species agrees with Smrz and Kocourková (1999) who studied Parmelia sulcata Taylor and Hypogymnia physodes (L.) Nyl. in the Czech Republic. The lack of difference we observed could be due to the similar growth forms and chemical composition of the species we studied. Other species of foliose lichen may be more difficult to sample because they are less common or do not form mats of 100% cover. Supplemental collections in June 2005 of mites on Lobaria pulmonaria (L.) Hoffm., a loosely attached foliose lichen with broad ascending lobes, yielded an additional species, Megeremaeus hylaius Behan-Pelletier, indicating that sampling of additional lichen species could reveal more mite taxa. Because mites are mobile and lichens can form complex matrices of mixed microhabitats, 434 Northeastern Naturalist Vol. 14, No. 3 future work might sample entire branches and correlate mite community composition with lichen community composition using methods such as canonical correlation analysis. Our MRPP and indicator species analyses of the substrates (Tables 4 and 5) indicate that crustose lichen and bark faunas are discernible from those of foliose lichens. Indicator species are associated with substrates (Table 5). Although some species were indicative of a particular substrate (Table 5), we infer that differences in mite communities are largely attributable to differing relative abundance of mite species. Estimates of arboreal mite population density previously have been reported per unit area without regard to particular arboreal substrates (Lamoncha and Crossley 1998, Wunderle 1992) or by 100 g dry mass lichen (Lindo and Winchester 2006). We cannot easily compare our estimates to those of previous studies because we did not sample every available substrate. Like others (André 1985; Fröberg et al. 2003), we found that mite community composition and abundance differ among particular substrate categories. Therefore, if lichen cover varies with tree size or among tree species, total mite densities would depend upon the diameter distribution and species composition of the stand. Research by Root and McGee (Root et al., in press) suggests that lichen cover is generally much lower on smaller trees. We therefore expect that stands comprised primarily of small trees would support lower mite densities than stands with many large trees. Mite community composition on large sugar maples differed significantly between old-growth and reserve shelterwood stands. Previous work has suggested two ecological groups of arboreal mites: generalist wandering species with broad dispersal, and specialized arboreal species with narrow dispersal (Winchester et al. 1999). If this hypothesis holds, the old growth indicator Platyliodes sp. 1, a member of a generally arboreal genus (Norton and Behan-Pelletier, in press), may be a poor disperser. Anachipteria magnilamellata, the other indicator of large trees in old growth, has not been studied extensively, but all collections known to us are from arboreal habitats. However, it is also possible that these species are particularly sensitive to alterations in habitat conditions. Reserve-shelterwood indicators Anachipteria n. sp. and Caleremaeus sp. 1 have not been described, and therefore, there is no obvious reason why either of these taxa should be a better disperser or more adapted to live in open-canopy stands. The apparent differences between old growth and reserve shelterwood contrast with Erdmann et al. (in press), who found no significant differences among mite communities in variously managed even-aged and virgin oak stands in Poland. However, both studies have small sample sizes and thus low power to detect differences. Further work with more intensive sampling in northeast North America should examine whether the differences we observed have ecological significance. Observations of live oribatid mites on several lichen specimens collected in 2005 revealed numerous occurrences of soredia caught in 2007 H.T. Root, G.G. McGee, and R.A. Norton 435 notogastral setae, particularly by Phauloppia banksi. If mites are potential dispersers of lichen spores (Meier et al. 2002), their dispersal abilities and association with uncut stands may be meaningful for lichen ecology; for example, it may contribute to the shared pattern of oribatid mites (Winchester et al. 1999) and epiphytic lichens (Hilmo and Sastad 2001, Keon and Muir 2002, Sillett et al. 2000) in which poorly dispersing species are restricted to ancient forests. Conclusions These preliminary data provide a first faunal description of mites present in canopies of sugar maple in the Adirondack Mountains of New York. In 0.26 m2 of substrate sampled, 877 individuals were collected and 24 oribatid mite taxa were represented, including at least three undescribed species. We demonstrated that foliose lichens provide habitats for a greater number of mites and a different community composition relative to crustose lichen or bark. Although old growth and reserve shelterwoods differ significantly in mite species composition, the reasons for, and ecological significance, of this difference are unknown. Further sampling of arboreal habitats could reveal other new species and facilitate understanding of the biology of mites, their life cycle and resource requirements, their potential contribution to arboreal trophic systems, and their interactions with epiphyte communities. Acknowledgments This work was part of a project funded by the Northeastern States Research Cooperative in conjunction with the University of Vermont and the USDA. Howard Prescott and Samuel Urffer assisted in field work and extractions. Robin Kimmerer, Alexander Weir, and Ralph Nyland reviewed earlier drafts. The manuscript benefited from review by two anonymous referees, to whom we are grateful. Literature Cited André, H.M. 1984. Notes on the ecology of corticolous epiphyte dwellers: Oribatida. Acarologia 25:385–395. André, H.M. 1985. Associations between corticolous microarthropod communities and epiphytic cover on bark. Holarctic Ecology 8:113–119. André, H.M., C. Bolly, and P. Lebrun. 1982. Monitoring and mapping air pollution through an animal indicator: A new and quick method. Journal of Applied Ecololgy 19:107–111. Behan-Pelletier, V.M. 1999. Oribatid mite biodiversity in agroecosystems: Role for bioindication. Agriculture Ecosystems and Environment 74(1–3):411–423. Behan-Pelletier, V.M., and D.E. Walter. 2000. Biodiversity of oribatid mites (Acari: Oribatida) in tree canopies and litter. Pp. 187–202, In D.C. Coleman and P.F. Hendrix (Eds.). Invertebrates as Webmasters in Ecosystems. CABI Publications, Wallingford, UK. 436 Northeastern Naturalist Vol. 14, No. 3 Behan-Pelletier, V.M., and N.N. Winchester. 1998. Arboreal oribatid mite diversity: Colonizing the canopy. Applied Soil Ecology 9:45–51. Braun, E.L. 1950. Deciduous Forests of Eastern North America. Hafner Publishing Company, New York, NY. 596 pp. Bray, J.R., and J.T. Curtis. 1957. An ordination of the upland forest communities in southern Wisconsin. Ecological Monographs 27:325–349. Cameron, R.P. 2002. Habitat associations of epiphytic lichens in managed and unmanaged forest stands in Nova Scotia. Northeastern Naturalist 9:27–46. Carroll, G.C. 1980. Forest canopies: complex and independent subsystems. Pp. 87– 108, In R.H. Waring (Ed.). Forests: Fresh Perspectives From Ecosystem Analysis. Oregon State University, Corvallis, OR. Dindal, D.L. (Ed.). 1990. Soil Biology Guide. Wiley and Sons, New York, NY. 1349 pp. Erdman, G., A. Floren, K.E. Lisenmair, S. Scheu, and M. Maraun. In press. Little effect of forest age on oribatid mites on the bark of trees. Pedobiologia. Forrester, J.A., G.G. McGee, and M.J. Mitchell. 2003. Effects of beech bark disease on aboveground biomass and species composition in a mature northern hardwood forest, 1985–2000. Journal of the Torrey Botanical Society 130:70–78. Fröberg, L., T. Solhøy, A. Baur, and B. Baur. 2003. Lichen specificity of oribatid mites (Acari: Oribatida) on limestone walls in the Great Alvar of Öland, Sweden. Entomologisk Tidskrift 124:177–182. Hilmo, O., and S.M. Sastad. 2001. Colonization of old-forest lichens in a young and an old boreal Picea abies forest: An experimental approach. Biological Conservation 102:251–259. Keon, D.B., and P.S. Muir. 2002. Growth of Usnea longissima across a variety of habitats in the Oregon Coast Range. The Bryologist 105:233–242. Krantz, G.W. 1978. A Manual of Acarology. Oregon State University, Corvallis, OR. 509 pp. Kruskal, J.B. 1964. Nonmetric multidimensional scaling: A numerical method. Psychometrika 29:115–129. Lamoncha, K.L., and D.A. Crossley, Jr. 1998. Oribatid mite diversity along an elevation gradient in a southeastern Appalachian forest. Pedobiologia 42:43–55. Lindo, Z., and N.N. Winchester. 2006. A comparison of microarthropod assemblages with emphasis on oribatid mites in canopy suspended soils and forest floors associated with ancient western redcedar trees. Pedobiologia 50:31–41. Marshall, V.G., R.M. Reeves, and R.A. Norton. 1987. Catalogue of the Oribatida (Acari) of continental United States and Canada. Memoirs of the Entomological Society of Canada 139:1–418. McCune, B. 1993. Gradients in epiphyte biomass in three Pseudotsuga-Tsuga forests of different ages in western Oregon and Washington. The Bryologist 96:405–411. McCune, B., and J.B. Grace. 2002. Analysis of Ecological Communities. MjM Software, Gleneden Beach, OR. 300 pp. McCune, B., and M.J. Mefford. 1999. PC-ORD Version 4.0. MjM Software, Gleneden Beach, OR. Meier, F.A., S. Scherrer, and R. Honegger. 2002. Faecal pellets of lichenivorous mites contain viable cells of the lichen-forming ascomycete Xanthoria parietina and its green algal photobiont, Trebouxia arboricola. Biological Journal of the Linnean Society 76:259–268. 2007 H.T. Root, G.G. McGee, and R.A. Norton 437 Nicolai, V. 1986. The bark of trees: Thermal properties, microclimate, and fauna. Oecologia 69:421–430. Nicolai, V. 1993. The arthropod fauna on the bark of deciduous and coniferous trees in a mixed forest of the Itasca State Park, MN. Spixiana 16:61–69. Norton, R.A., and Behan-Pelletier, V.M. Oribatida. In press. In G.W. Krantz and D.E. Walter (Eds.). A Manual of Acarology. Texas Tech University Press, Lubbock, TX. Nyland, R.D. 2002. Silviculture: Concepts and Applications. McGraw-Hill, New York, NY. Pettersson, R.B., J.P. Ball, K.-A. Renhorn, P.-A. Esseen, and K. Sjöberg. 1995. Invertebrate communities in boreal forest canopies as influenced by forestry and lichens with implications for passerine birds. Biological Conservation 74:57–63. Prinzing, A. 2005. Corticolous arthropods under climatic fluctuations: Compensation is more important than migration. Ecography 28:17–28. Reeves, R.M. 1988. Distribution and habitat comparisons for Carabodes collected from conifer branches with descriptions of brevis Banks and higginsi sp. n. (Acari: Oribatida: Carabodidae). Proceedings of the Entomological Society of Washington 90:373–392. Reeves, R.M., and V.M. Behan-Pelletier. 1998. The genus Carabodes (Acari: Oribatida: Carabodidae) of North America, with descriptions of new western species. Canadian Journal of Zoology 76:1898–1921. Richardson, D.H.S., and R.P. Cameron. 2004. Cyanolichens: Their response to pollution and possible management strategies for their conservation in northeastern North America. Northeastern Naturalist 11:1–22. Root, H.T., G.G. McGee, and R.D. Nyland. In press. Effects of two silvicultural management regimes with large tree retention on epiphytic lichen communities in Adirondack northern hardwoods, New York, USA. Canadian Journal of Forest Research. SAS Institute, 2005. SAS. SAS Institute, Cary, NC, Seniczak, S., J. Dabrowski, A. Klimek, and S. Kaczmarek. 1996. The mites associated with young Scots pine forests polluted by a copper smelting works in Glogow, Poland. Pp. 573–574, In R. Mitchell, D.J. Horn, G.R. Needham, and W.C. Welbourn (Eds.). Acarology. IX. Vol. 1. Proceedings, Ohio Biological Survey, Columbus, OH. Seyd, E.L., and M.R.D. Seaward. 1984. The association of oribatid mites and lichens. Zoological Journal of the Linnean Society 80:369–420. Shigo, A.L. 1972. The beech bark disease today in the northeastern United States. Journal of Forestry 70:286–289. Sillett, S.C., B. McCune, J.E. Peck, T.R. Rambo, and A. Ruchty. 2000. Dispersal limitations of epiphytic lichens result in species dependent on old-growth forests. Ecological Applications 10:789–799. Smrz, J. and J. Kocourková. 1999. Mite communities of two epiphytic lichen species (Hypogymnia physodes and Parmelia sulcata) in the Czech Republic. Pedobiologia 43:385–390. Stubbs, C.S. 1989. Patterns of distribution and abundance of corticolous lichens and their invertebrate associations on Quercus rubra in Maine. The Bryologist 92:453–460. 438 Northeastern Naturalist Vol. 14, No. 3 Travé, J. 1963. Ecologie et biologie des Oribates (Acariens) saxicoles et arboricoles. Vie et Milieu Supplement 14:1–267. van Straalen, N.M. 1998. Evaluation of bioindicator systems derived from soil arthropod communities. Applied Soil Ecology 9:429–437. Walter, D.E., and H.C. Proctor. 1999. Mites: Ecology, Evolution, and Behaviour. CABI International, Wallingford, UK. Winchester, N.N., V.M. Behan-Pelletier, and R.A. Ring. 1999. Arboreal specificity, diversity, and abundance of canopy-dwelling oribatid mites (Acari: Oribatida). Pedobiologia 43:391–400. Wunderle, I. 1992. Die Oribatiden-Gemeinschaften (Acari) der verschiedenen Habitate eines Buchenwaldes. Carolinea 50:79–144.