Northeastern Naturalist 214 G.G. McGee, M.E. Cardon, and D.H. Kiernan 22001199 NORTHEASTERN NATURALIST 2V6(o1l). :2261,4 N–2o3. 51 Variation in Acer saccharum Marshall (Sugar Maple) Bark and Stemflow Characteristics: Implications for Epiphytic Bryophyte Communities Gregory G. McGee1,*, Megan E. Cardon1, and Diane H. Kiernan2 Abstract - Past studies have demonstrated successions of epiphyte communities on trees of progressively larger diameters, suggesting the existence of temporal resource-gradients associated with aging/growing phorophyte hosts. The objective of this study was to identify possible resource gradients associated with the diameter/age of Acer saccharum (Sugar Maple) in northern New York. We determined epiphytic bryophyte cover by species on 102 Sugar Maples (min–max = 11–84 cm diameter at breast height) from 12 Adirondack northern hardwood forest stands. We extracted a 12.6-cm2 bark sample from each tree to analyze in the laboratory for moisture-holding capacity, surface-moisture availability, drying rates, and leachate cation and nitrogen concentrations. We collected throughfall and stemflow from 15 trees (19–79 cm dbh) at a separate site over the course of a growing season and analyzed samples for cation and nitrogen concentrations. Bark mass per unit surface area (g cm-2) was positively correlated with tree diameter, reflecting increasing bark thickness with age. Bark moisture-holding capacity (H2O as % dry mass) was independent of tree diameter, but bark surface-moisture availability (g H2O cm-2 bark) increased with diameter as a result of thickening bark. Bark drying rates were negatively correlated with bark mass (thickness). Cation (Ca2+, Mg2+, K+) concentrations in bark leachate were all positively correlated with tree diameter, but NH4 + and DON concentrations varied independently of tree diameter, and NO3 - concentrations were typically below detection limits. Stemflow became enriched 10- to 20-fold with dissolved cations but not with dissolved nitrogen. Percent cover of several mesophytic and calciphilic epiphytes (e.g., Anomodon rugelii, Brachythecium laetum, Neckera pennata, and Porella platyphylla) were positively correlated with cation concentration in bark leachate, bark thickness, and moisture availability, and negatively correlated with bark drying rate. The results of this study are consistent with hypotheses that increased moisture and nutrient availability and slower drying rates of bark on large-diameter trees may account for increasing total cover and species richness of bryophtyes and increasing dominance of mesophytic and calciphilic bryophytes on larger trees. We extend McCune’s similar gradient hypothesis with an analogous set of nutrient-based gradients, and offer an alternative mechanism for McCune’s original time-based moisture gradient. Introduction Epiphytic bryophyte communities are influenced by complex gradients of substrate moisture, atmospheric humidity, and nutrient and light availability that act simultaneously at regional, local, and micro-environmental scales. Regionally, 1Department of Environmental and Forest Biology, State University of New York College of Environmental Science and Forestry, Syracuse, NY 13210. 2Department of Forest and Natural Resource Management, State University of New York College of Environmental Science and Forestry, Syracuse, NY 13210. *Corresponding author - ggmcgee@esf.edu. Manuscript Editor: David Richardson Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 215 epiphyte communities have long been considered to be affected by temperature and evapotranspirational gradients (e.g., Phillips 1951, Slack 1976). At more local (forest-stand level) scales, epiphyte community composition appears to be influenced by the manner by which slope position and aspect affect relative humidity (Barkman 1958:69, Gimingham and Birse 1957, McCune 1993), how canopy disturbance and understory vegetation affect light, temperature and relative humidity (Boudreault et al. 2008, Gustafsson and Eriksson 1995, Halpern et al. 2014, Király and Ódor 2010, McCune 1993), and how soil chemistry interacts with bark chemistry to influence nutrient availability for epiphytes (Bates 1992, Boudreault et al. 2008, Gustafsson and Eriksson 1995). At the microenvironmental scale, interspecific variation in physical bark features (thickness, depth, moisture-holding capacity) and chemical conditions (pH, nutrient availability) of different phorophyte host species determine (or have been invoked to explain) epiphyte host preferences (Barkman 1958; Bates 1992; Billings and Drew 1938; Cain and Sharp 1938; Cleavitt et al. 2009; Culberson 1955; Hale 1955; Király and Ódor 2010; McGee and Kimmerer 2002; Mezaka et al. 2008, 2012; Ódor et al. 2014; Slack 1976; Studlar 1982a; Szövényi et al. 2004). Further, the vertical distributions of epiphytes on trees within a forest stand are associated with vertical gradients of humidity and light (Billings and Drew 1938, Hosokawa et al. 1964, McCune 1993, Szovenyi et al. 2004). Xerophytic and light-demanding species occur higher on tree boles, while mesophytic and shade-tolerant species occur near the ground in zones of higher humidity, and away from desiccating air currents (Hosokawa and Odani 1957, Hosokawa et al. 1964). However, in addition to the predictable variation that epiphytic bryophytes display across landscapes and hillslopes, among phorophyte species, and along the vertical gradients of individual phorophytes, a growing body of research has also demonstrated increased richness (Hazell et al. 1998, Mezaka et al. 2008), abundance of red-listed species (Fritz et al. 2009, Mezaka et al. 2008, Snäll et al. 2004), and turnover in overall epiphyte community composition in association with increasing phorophyte host diameter (Aude and Poulsen 2000, Boudreault et al. 2008, Fritz et al. 2009, Hazell et al. 1998, McGee and Kimmerer 2002, Studlar 1982b). For instance, an associated study in northern hardwood forests of northern New York demonstrated variation in epiphytic bryophyte communities that was consistently correlated with Acer saccharum Marsh. (Sugar Maple) diameter, regardless of stand disturbance or management history (McGee and Kimmerer 2002). Frequency and total cover of mesophytic epiphytes such as Brachythecium laetum (= B. oxycladon) (Bridel) Schimper in P. Bruch and W.P. Schimper, B. salebrosum (Hoffmann ex F. Weber & D. Mohr) Schimper in P. Bruch and W.P. Schimper, Dicranum viride (Sullivant & Lesquereux) Lindberg, Leucodon brachypus var. andrewsianus Crum & Anderson, Neckera pennata Hedwig, Plagiomnium ciliare (Müller Hal.) T.J. Koponen,, and Porella platyphylla (L.) Pfeiff., and calciphilic species such as Anomodon rugelii (Müll. Hal.) Keissl., A. attenuatus (Hedw.) Hüb., and Radula complanata (L.) Dumort. were positively Northeastern Naturalist 216 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 correlated with Sugar Maple diameter, while small-diameter Sugar Maples were dominated by xerophytic species such as Ulota crispa (Hedw.) Brid., Hypnum pallescens var. pallescens (Hrdw.) P. Beauv., Sciuro-hypnum reflexum (= B. reflexum) (Starke) Ignatov & Huttunen, Platygyrium repens (Brid.) Schimp., and Frullania eboracensis Gottsche (nomenclature follows Crum [1991] for liverworts, and Flora of North America [2007, 2014] for mosses; a single taxonomic treatment, P. platyphylla, is applied to P. platyphylla and P. platyphyloidea per James et al. [1998]). Therefore, just as multiple spatial resource-gradients appear to influence epiphyte composition at the landscape, stand, and microenvironmental scales, similar temporal resource gradients probably exist that may influence epiphyte community composition through time on a given phorophyte host. Such temporal environmental variation may be due to changing physical and chemical conditions that accompany the development of mature bark characteristics of many tree species such as Sugar Maple. To date, few studies have investigated intraspecific variation in the physical and chemical conditions of tree bark. Bates (1992) determined that bark pH and Mg concentration were positively correlated with diameter of Fraxinus excelsior L. (European Ash). Similarly, Andersson (1991) reported positive correlations between stemflow cation-concentration and diameter of Quercus robor L. (English Oak). However, Gustafsson and Eriksson (1995) found no relationship between bark chemistry and tree diameter during an investigation of epiphytic communities on Populus tremula L. (European Aspen), and Bourdreault et al. (2008) reported for Populus tremuloides Michx. (Quaking Aspen) no correlation between bark pH and diameter, a negative correlation between bark Ca- and Mg-content and diameter, and a positive correlation between bark K-content and diameter. Therefore, given the range of conclusions from the few past efforts to characterize changes in physical and chemical properties of phorophyte bark over time, our objective was to contribute to a greater understanding of temporal microenvironmental resource gradients associated with aging trees and their potential influences on epiphytic bryophyte community composition. Specifically, this study focused on variation in Sugar Maple bark that may account for the positive relationships between tree diameter and epiphytic bryophyte richness and mesophytic/calciphilic bryophyte abundance. We tested the following 4 hypotheses: (1) the moisture-holding capacity (water as percent dry mass of bark) and surface moisture availability (g water cm-2 bark) of the bark will increase, and bark drying rate (% water mass loss hour-1) will decrease with increasing Sugar Maple diameter; (2) bark leachate will become enriched in elemental nutrients (μmol L-1 NO3 -, NH4 +, DON, Ca2+, K+ and Mg2+) with increasing Sugar Maple diameter; (3) percent cover of mesophytic bryophytes will be positively correlated with bark surface moisture availability, which is, in turn, influenced by tree diameter; and (4) percent cover of calciphilic bryophytes will be positively correlated with bark leachate cation concentrations, which are, in turn, influenced by tree diameter. Finally, we used observations of stemflow chemistry from 15 Sugar Maple trees growing within a single forest stand to corroborate and contextualize our bark-chemistry data. Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 217 Methods Sugar Maple bark characteristics Study sites. We conducted field sampling for this study in Adirondack northern hardwood forests (as described by Braun 1950), which are dominated by Sugar Maple, Fagus grandifolia Ehrh. (American Beech), and Betula alleghaniensis Britton (Yellow Birch), with conifer associates including Tsuga canadensis (L.) Carr (Eastern Hemlock), and Picea rubens Sarg. (Red Spruce). Within the Adirondack region, northern hardwoods typically occur at elevations below 980 m (all field sites in this study were between 470 m and 770 m) and on loamy glacial till soils overlying granitic gneiss (Heimburger 1934). We established 12 study sites in 1996 in 4 replicate stands representing 3 stand histories. At each study site, we established two 0.1-ha plots to quantify overstory composition and structure (trees > 10.0 cm diameter at breast height [dbh]). McGee et al. (1999) and McGee and Kimmerer (2002) described disturbance histories, and stand structure and composition of the 3 stand types in detail, but brief descriptions follow. Old-growth sites were uneven-aged stands with average ages of dominant trees varying from 176 y to 212 y. Most sites were likely selectively harvested for only large-diameter Red Spruce prior to their incorporation into the Adirondack Park in the early 1890s. The old-growth stands averaged 55 trees ha-1 greater than 50 cm dbh. Basal areas averaged (± 1 SD) 33.7 ± 4.3 m2 ha-1, and total tree-surface area available for epiphyte establishment averaged 560 ± 17 m2 ha-1 (from 0 m to 1.5 m above the ground). Maturing sites were 90- to 100-y–old even-aged stands that regenerated from wildfires at the turn of the 20th century and have not been logged since establishment. Average stand diameters of the post-fire cohorts varied from 23 ± 10 cm to 25 ± 10 cm dbh. Increment coring confirmed that an average of 8 trees ha-1 (all ≥ 50 cm dbh) were residuals that survived the fires. Live basal areas averaged 29.1 ± 4.0 m2 ha-1, and total tree-surface area averaged 654 ± 113 m2 ha-1. Partially cut sites had received repeated, undefined, partial cuts for more than 100 years. Most recently they have been under 15- to 20-y cutting cycles, with maximum-diameter limits of 45–60 cm dbh and residual basal areas of ~16 m2 ha-1. The partially cut stands contained an average of 5 trees ha-1 greater than 50 cm dbh. Live basal areas averaged 18.2 ± 1.7 m2 ha-1, and total tree-surface area averaged 471 ± 57 m2 ha-1. Epiphyte field-sampling and cover estimation. Epiphyte communities at the 12 sites were initially sampled during 1996–1997, and overall epiphyte community composition across sites and stand types was described by McGee and Kimmerer (2002). The current study focused on only a subset of 102 Sugar Maples that were part of the initial 241-tree sample, and which originally included a wider variety of phorophyte hosts. We selected epiphyte sample trees through a stratified (by 10- cm–diameter classes) random sample of stems occurring on the 2 plots at each study site. We sampled 1 tree per diameter class, as available, on each plot (Table 1). Tree diameters varied from 11 cm to 84 cm dbh. Northeastern Naturalist 218 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 We employed line-intercept sampling to estimate percent bryophyte cover along 4 circumferential transects established on each of the 102 trees at 10 cm, 50 cm, 100 cm, and 150 cm above the ground surface. Determination of height-weighted percent cover (from 0 m to 1.5 m above the ground) for epiphytic bryophyte species are detailed in McGee and Kimmerer (2002). Briefly, we spaced line-intercepts at unequal height intervals along the tree bole in order to account for rapid change in total epiphyte cover and turnover in species composition at lower bole positions. However, since transects were spaced unequally and the tree circumferences differed at each height along each bole, the 4 transects accounted for unequal proportions of each tree’s available surface area. Therefore, the contribution of each transect to estimating the total cover of each species was weighted by the surface area of the cylinder it described using the following equation. d d % cover = Σ Pi([Areai] / [Σ Areai]) * 100, i = a i = a where i represents transects a–d (transect a at 1.5 m above the ground, b at 1.0 m, c at 0.5 m, and d at 0.2 m); Pi is the proportion of epiphyte cover on each transect and is defined as the number of 1-cm intervals containing epiphyte cover divided by the total length (cm) of the transect; and Areai = Li * HTi, with Li equaling the length of each transect i, and HTi being the vertical extent of the tree bole to which each intercept is applied (HTa = 50 cm, HTb = 50 cm, HTc = 30 cm, HTd = 20 cm). Note that transect d was placed at 10 cm above the ground, but we applied a height of 20 cm for the purpose of estimating height-weighted percent cover. Many basal epiphytes usually occurred above 10 cm but below 50 cm; thus, the height-weighted cover of basal epiphytes would be systematically underestimated by applying HTd = 10 cm. Therefore, we applied HTd = 20 to partially correct this sample bias. Bark sampling and laboratory analyses. In September 1997, we extracted bark samples from a randomly determined aspect of each tree 50 cm above the ground by hammering a sharpened 4.0-cm inside-diameter (12.6 cm2) steel pipe into the bark. Since we removed both inner and outer bark down to cambium, bark samples varied in thickness based on the condition of each tree. We carefully removed all Table 1. Summary of epiphyte-community and bark-chemistry sampling effort on 102 Sugar Maples, by stand type and diameter class, across 12 Adirondack northern hardwood forest study-sites. Diameter class Number of trees sampled (cm dbh) Old growth Maturing Partially cut Total 10.0–19.9 4 7 8 19 20.0–29.9 6 7 7 20 30.0–39.9 6 7 8 21 40.0–49.9 5 6 6 17 50.0–59.9 7 0 3 10 60.0–69.9 5 2 0 7 70.0–79.9 4 1 0 5 80.0–89.9 3 0 0 3 Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 219 lichens and bryophytes and stored each bark sample in a waxed paper envelope at 4 °C prior to analysis. We determined the moisture-holding capacity (g water g-1 dry wt bark, expressed as percent) and surface-moisture availability (g water cm-2 bark) as follows: the bark samples were soaked in 50 ml of deionized water for 24 h at 4 °C; drained for 15 min; weighed in their saturated state, air-dried in the laboratory and reweighed at 1, 2, 4, 9, 16, 25, 36, 49, and 64 h; and oven-dried at 40 °C for 48 h to obtain dry mass. We defined the drying rate of each bark as the slope of the least-squares regression equation describing the relationship between the percentage of the initial saturated bark water content remaining and time (h) for the drying curve to become asymptotic. Each of the 102 drying-curve asymptotes was defined by applying successive linear regressions using a spline model (Freund and Littell 1991, Littell et al. 1991). It was evident by observing the drying curves that they became asymptotic at 4–16 h. Therefore, we applied successive spline regressions while changing the spline point within this timeframe (i.e., 4, 9, 16 h). We used the spline that maximized the amount of variation accounted for by the regression to define the asymptote. We filtered bark extracts (leachate remaining from soaking bark) through a 0.8-μm Supor-800 membrane filter (Pall Corporation, Port Washington, NY). We determined concentrations (μmol L-1) of Ca2+, K+, and Mg2+ in the leachate using inductively coupled plasma atomic-emissions spectroscopy; nitrate (NO3 -) and ammonium (NH4 +) concentrations using an ion chromatograph (Small et al. 1975) and phenate colorimetry autoanalyzer (US EPA 1983); and total dissolved nitrogen (TDN) on a Bran Luebbe AA3 auto analyzer after persulfate oxidation (Ameel et al. 1993) and dissolved organic nitrogen (DON) by difference between TDN and the inorganic N fractions (NO3 - and NH4 +). Analyses. We used nonmetric multidimensional scaling (NMS) ordination to detect and visualize correlations between bark characteristics and epiphytic bryophyte community composition. The primary NMS ordination matrix consisted of height-weighted percent cover for 11 species on 102 trees. We limited the ordination analysis to only those species that were determined in an earlier study (McGee and Kimmerer 2002) to characterize epiphyte communities on smalldiameter (S. reflexum, H. pallescens, Ulota crispa, F. eboracensis and P. repens) and large-diameter (A. attenuatus, A. rugelii, B. laetum, L. brachypus, N. pennata, and P. platyphylla) Sugar Maples. The secondary ordination matrix consisted of 9 quantitative chemical and physical characteristics of the102 Sugar Maple bark samples (bark leachate Ca2+, K+, Mg2+, NH4 +, and DON concentrations; bark drying rate; dry mass; moisture availability; and tree dbh). We conducted the ordination in PC-ORD using Sørensen distances in slow and thorough autopilot mode from a random starting configuration (McCune and Mefford 2011). The cutoff R2 value for including a descriptor-variable vector in the joint plot was set to 0.20. We used regression analyses to determine univariate relationships between tree diameter, physical bark characteristics, and nutrient concentrations (n = 102). Data on drying rate and dry mass were log10-transformed to linearize their relationships with tree diameter prior to analyses. We used correlation analyses Northeastern Naturalist 220 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 to describe the significance and strength of the most-important relationships detected by the NMS ordination between bark characteristics and height-weighted percent cover of the epiphytes. Sugar Maple stemflow Study site. We conducted the stemflow study in an uneven-aged northern hardwood forest at the SUNY-ESF Huntington Wildlife Research Station, Newcomb, Essex County, NY (74°15'30''W, 44°00'30''N). The research station was situated roughly at the geographic center of the other 12 field-sampling sites. The study site was located on a northwest-facing toe slope. The soil type at the site was a loamyskeletal, mixed, frigid Typic Haplorthod (Somers 1986). We selected from within a 0.5-ha area a stratified (by 10-cm–diameter classes) random sample of 15 canopydominant and codominant Sugar Maples, varying in diameter from 19 cm dbh to79 cm dbh. Field sampling. We constructed stemflow collectors from 1.5-cm outside-diameter Tygon tubing. The top quarter of the tubing was removed to make a trough that we stapled in a spiral around each tree and sealed with silicone caulk. We emptied the tubing into 10-L plastic buckets stored in covered 135-L plastic containers. We collected the throughfall under each tree in a 2-L plastic beaker fitted with a 2-mm fiberglass-mesh filter, with 1 of these throughfall collectors randomly placed under the canopy of each tree and randomly repositioned after each precipitation event. We used a pair of 2-L plastic beakers fitted with a fiberglass mesh filter to collect bulk precipitation in the nearest clearing, ~2 km from the stemflow-collection area. We collected all stemflow, throughfall, and precipitation samples (60–120 ml) within 24 h after 13 rain events between 9 May and 12 September 1997 and stored them at 4 °C until analyzed. It should be noted that not all trees produced sufficient stemflow volume to conduct all analyses for all events. The water-collection vessels and tubing were rinsed with deionized water following each collection. We analyzed nitrate and NH4 + with an ion chromatograph (Small et al. 1975) and phenate colorimetry autoanalyzer (USEPA 1983), and cations (K+, Ca2+, Mg2+) by inductively coupled plasma atomic-emissions spectroscopy. For the purpose of our analyses, we applied concentrations of zero to all measurements that were below analytical detection limits. Results Bark physical characteristics The NMS ordination of Sugar Maple epiphyte community composition indicated that the percent cover of a number of species (A. rugelii, A. attenuatus, P. platyphylla, N. pennata L. brachypus, and B. laetum) was positively correlated with tree diameter and several physical and chemical bark characteristics including dry mass, moisture availability, and Ca2+ and Mg2+ concentrations in leachate (Fig. 1). Other species (F. eboracensis, U. crispa, S. reflexum, P. repens, and H. pallescens) exhibited negative correlations with tree diameter and values of various chemical and physical bark characteristics. While the old-growth epiphyte communities Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 221 appeared to separate from the communities of the maturing and partially cut stands along the second NMS ordination axis, this difference probably reflects the greater relative proportion of large trees sampled in the old-growth stands and of smaller trees in the maturing and partially cut stands (Table 1). Diameters of the 102 Sugar Maples averaged (± 1 SD) 37.3 ± 18.1 cm dbh, and ranged from 11 dbh to 84 cm dbh. Dry mass of the 12.6-cm2 bark samples varied from 0.12 g cm-2 to 0.92 g cm-2, averaged 0.34 ± 0.17 g cm-2, and was positively correlated with tree diameter (Fig. 2a; P < 0.001, R2 = 0.35). A log10-transformed regression model accounted for a slightly greater level of variation in the data Figure 1. NMS ordination of epiphytic bryophyte communities and associated bark characteristics on Sugar Maples (n = 102) in Adirondack Park, NY. Species are: Anomodon rugelii (Anorug), Brachythecium laetum (Bralae), Frullania eboracensis (Fruebo), Hypnum pallescens var. pallescens (Hyppal), Leucodon brachypus var. andrewsianus (Leubra), Neckera pennata (Necpen), Platygyrium repens (Plarep), Porella platyphylla (Porpla), Sciuro-hypnum reflexum (Sciref), and Ulota crispa (Ulocri). Northeastern Naturalist 222 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 compared to a linear model, and more adequately described the asymptotic trend in bark dry mass, which tended to level off at ~60–70 cm dbh. At saturation, bark moisture content (mass of water as % bark dry weight) varied from 48% to195% and averaged 86 ± 26%, but did not vary in relation to tree diameter (Fig. 2b; P = 0.25, R2 = 0.02). Bark surface-moisture availability varied from 0.09 g to 0.67 g Figure 2. Relationships between Sugar Maple diameter and (a) bark dry mass, (b) bark moisture content, (c) bark moisture availa b i l i t y, and (d) bark drying rate. Regressions summarized in panels (a) and (d) exhibited superior fits with l o g a r i t h m i c models. Data are presented with arithmetic scales. Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 223 water cm-2 bark, averaged 0.28 ± 0.13 g water cm-2 bark, and increased with increasing tree diameter (Fig. 2c; P < 0.001, R2 = 0.45). After 4–9 h of drying under laboratory conditions, the mass of water-saturated bark became asymptotic (i.e., R2 values of spline regressions peaked [R2 min = 0.76, max = 0.99, average = 0.96] at 4 h of drying for all 102 samples). Pre-asymptotic drying rates (loss of mass h-1 as percentage of initial water mass in saturated bark) varied from ~10% h-1 to 22% h-1 and averaged 16.9 ± 2.6% h-1. Bark drying rates were negatively correlated with log10-transformed tree dbh (P < 0.001, R2 = 0.32), with bark of small-diameter trees drying at a rate ~1.3-times faster than bark of large-diameter trees (Fig. 2d). Bark chemical characteristics Average (± 1 SD) cation concentrations of bark leachate (μmol L-1) were 201 ± 170 for Ca2+, 76 ± 60 for Mg2+, and 879 ± 817 for K+. Concentrations of Ca2+ (P < 0.001, R2 = 0.41), Mg2+ (P < 0.001, R2 = 0.42), and K+ (P < 0.001, R2 = 0.24) in bark leachate all increased ~5-fold from the smallest to largest diameter trees (Fig. 3a–c). Nitrogen in the bark leachate was dominated by the DON fraction (29 ± 38 μmol L-1). Nitrate concentrations averaged 0.4 ± 1.9 μmol L-1, with only 3 of the 102 bark samples yielding NO3 - above detection limits. Ammonium concentrations averaged 5 ± 10 μmol L-1, with 32 bark samples yielding NH4 + above detection limits. Unlike the cation nutrients, concentrations of DON (P = 0.60), NH4 + (P = 0.58), and NO3 - (P = 0.25) in bark leachate exhibited no correlations with tree diameter. Epiphyte–bark characteristic relationships In general, when correlations existed between percent cover of epiphyte species and the various chemical and physical bark variables, species in the large-tree epiphyte guild consistently exhibited positive correlations with bark nutrient and moisture variables (Table 2, Fig. 1). Further, as a group, the percent cover of species in the large-tree guild tended to have frequent and strongly significant correlations with the variables we measured, while cover of those in the small-tree guild tended to be independent of or more weakly negatively correlated with the bark variables (Table 2). The percent cover of the 5 large-tree epiphyte species were generally positively and highly correlated with most or all of the 3 measured mineral-cation concentrations in bark leachate, positively correlated with bark surface-moisture availability, and negatively correlated with bark drying rate (Table 2). The percent cover of no epiphyte species was correlated with DON (which was the most abundant nitrogen fraction), but 2 taxa (A. rugelii and P. platyphylla) exhibited positive correlations with NH4 +, which consistently occurred at concentrations lower than those of DON. Stemflow chemistry Over the duration of the study, bulk precipitation averaged (± 1SD) 9 ± 2 μmol L-1 Ca2+, 2 ± 1 Mg2+, 10 ± 2 K+, 27 ± 6 NO3 -, and 24 ± 8 NH4 +. As precipitation passed through the Sugar Maple canopies, it was enriched 3- to 7-fold in Ca2+, Mg2+, Northeastern Naturalist 224 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 and K+, but not in NO3 - and NH4 + (Fig. 4). Stemflow was then enriched again in cations 10- to 20-fold over throughfall and averaged 265 ± 192, 1167 ± 824, and 87 ± 65 μmol L-1 of Ca+2, K+, and Mg+2 respectively. There was no additional enrichment of stemflow with NO3 - and NH4 +. Discussion Bark physical conditions This study elucidated changes in the physical characteristics of Sugar Maple bark that occur as trees age and grow, and which may account in part for the Figure 3. Relationship between Sugar Maple diameter and (a) calcium, (b) magnesium, and (c) potassium concentrations of bark extract. Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 225 succession in epiphytic bryophyte communities observed in the literature. Bark mass per unit area increased 4- to 5-fold from the smallest to largest trees we studied (11–84 cm dbh), and while we did not directly measure the thickness of the unevenly platy and fissured bark of these trees, the increasing mass per standardized area of bark directly reflects the overall thickening of Sugar Maple bark with increasing tree diameter. The volume of water absorbed as a percentage of bark dry weight did not vary as a function of tree diameter, which suggests that no physical changes occur during the development of mature Sugar Maple bark that influence moisture absorption (e.g., changes in density or pore space). Rather, moisture retention in Sugar Maple bark appears to be a simple function of its thickness, resulting in the capacity for larger trees to hold 4-times more moisture per unit surface area than the smallest trees we studied. Furthermore, stored water evaporated under laboratory conditions from the thin bark of small trees at a rate 1.3-times faster than from thick bark of large trees. Thus, our data indicate that bark of large Sugar Maple trees retains a greater volume of water per unit of surface area, and dries more slowly than that of small-diameter trees, and this may be of consequence to Table 2. Correlations of bark-leachate nutrient concentrations and bark physical properties with height-weighted mean percent cover of individual large- and small-tree epiphytic bryophyte species. Pearson correlation coefficients (R) and P-values (P) are shown (* indicates significant interactions; P-value ≤ 0.05). A Bonferonni adjusted P-value < 0.0007 is required to control for experiment-wise error rate for 80 individual correlation analyses using α = 0.05. Moist. avail. = moisture availability. Bark leachate nutrient concentration Bark physical properties Org. Drying Dry Moist. Species Mg2+ Ca2+ K+ NH4+ N rate mass avail. Large-tree species Anomodon rugelii R 0.37* 0.39* 0.35* 0.34* -0.02 -0.28* 0.39* 0.34* P 0.000* 0.000* 0.000* 0.001* 0.81 0.004* 0.000* 0.000* Brachythecium laetum R 0.48* 0.44* 0.28* -0.06 -0.07 -0.36* 0.42* 0.48* P 0.000* 0.000* 0.004* 0.49 0.48 0.000* 0.000* 0.000* Leucodon brachypus R 0.22* 0.15 0.20* 0.03 0.03 -0.19 0.08 0.20* P 0.020* 0.130 0.040* 0.750 0.800 0.060 0.450 0.040* Neckera pennata R 0.22* 0.29* 0.14 0.03 -0.06 -0.22* 0.26* 0.28* P 0.020* 0.003* 0.170 0.79 0.570 0.030* 0.008* 0.005* Porella platyphylla R 0.54* 0.52* 0.54* 0.20* -0.28 -0.45* 0.52* 0.59* P 0.000* 0.000* 0.000* 0.040* 0.790 0.000* 0.000* 0.000* Small-tree species Frullania eboracensis R -0.12 -0.06 -0.03 -0.09 -0.12 -0.09 -0.05 -0.04 P 0.220 0.530 0.800 0.360 0.250 0.370 0.600 0.680 Hypnum pallescens R -0.08 -0.63 -0.11 0.02 0.03 0.23* -0.20* -0.18 P 0.410 0.530 0.300 0.840 0.780 0.020* 0.040* 0.070 Platygyrium repens R -0.26* -0.18* -0.19* -0.10 0.12 0.14 -0.23* -0.17 P 0.008* 0.050* 0.050* 0.300 0.240 0.170 0.020* 0.080 Sciuro-hypnum reflexum R -0.18 -0.18 -0.17 -0.15 -0.02 0.13 -0.14 -0.19 P 0.070 0.070 0.100 0.140 0.840 0.190 0.160 0.060 Ulota crispa R -0.17 -0.13 -0.26* -0.05 -0.04 0.14 -0.26* -0.24* P 0.090 0.200 0.010* 0.600 0.700 0.150 0.010* 0.020* Northeastern Naturalist 226 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 mesophytic epiphytes that utilize moisture reserves in bark. The additional moisture reserves and slower drying rate may extend the period of hydration and active metabolism for bryophytes between precipitation events. While several past studies have described interspecific differences in bark moisture-holding capacity (Billings and Drew 1938, Culberson 1955, Everhart et al. 2009, Hale 1955, Levia and Herwitz 2005, Studlar 1982a), few have reported on any intraspecific variation, particularly with regard to stem age or size. Everhart et al. (2009) observed thicker Platanus occidentalis L. (American Sycamore) bark, which occurred lower on trees, absorbed greater volumes of water than thin bark higher on trees, but did not report findings on a per mass basis. Hoffman and Boe (1977) reported a small (less than 2%) but significant decline in bark moisture-holding capacity (g water • g dry bark-1) with increasing diameter of Populus deltoides W. Bartram ex Marshall (Eastern Cottonwood) varying from ~60 cm to 90 cm dbh. Although the thicker Sugar Maple bark samples from large-diameter trees held more moisture and dried out more slowly than thinner samples from small-diameter trees, the relationships observed in this study do not offer direct evidence that bark serves as a physiologically meaningful capacitor of moisture for epiphytes (i.e., absorbing moisture during precipitation events while slowly releasing it over extended time periods afterward). Recent technological advances are making it possible to monitor vapor pressure at the small spatial scale necessary to determine whether bryophytes actually acquire physiologically meaningful levels of supplemental moisture from their substrate. Techniques pioneered by Haughian and Frego (2017a, b) for epixylic bryophyte communities could be modified and applied to Figure 4. Average ±1 SD concentration of inorganic nutrients in bulk precipitation, throughfall, and stemflow associated with 15 Sugar Maple trees in an Adirondack northern hardwood forest over 13 precipitation events during the 1997 growing season, and in bark leachate of 102 Sugar Maples. Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 227 better quantify vertical gradients of vapor pressure associated with phorophyte hosts of varying species, diameters, and local canopy disturbances. Bark chemical conditions Limitations in mineral nutrient availability are consequential for many bryophytes. Due to the level of cellular damage that occurs following rehydration, bryophytes are limited in their tolerance to repeated or extended periods of desiccation, especially in the absence of nutrient-rich rehydration solution. It is during the initial period of precipitation events when previously desiccated epiphytic bryophytes are most vulnerable to nutrient leaching through disrupted plasma and organelle membranes (Gupta 1977, Proctor 1982), and Brown (1982) hypothesized that membrane structures of calciphilic bryophytes require greater quantities of calcium to maintain integrity after desiccation. Not only does the duration and intensity of drying affect recovery, so does the rate of drying (Proctor et al. 2007). Rapid desiccation leads to extreme degradation of chloroplasts and mitochondria, resulting in additional loss of photosynthetic activity and disruption of ATP synthesis (Richardson 1981). Therefore, high nutrient concentrations are vital to rehydrating bryophytes in order to lower diffusion gradients and conserve cytoplasmic nutrients. To date, few studies have considered elemental nutrient concentration of tree bark or given particular consideration to intraspecific variation associated with age/ girth of trees or site conditions. Bates (1992) and Boudreault et al. (2008) reported total cation concentrations per gram dry weight of bark (using acid digestions of bark samples) for Quercus petraea (Matt.) Liebl. and European Ash in Scotland, and Quaking Aspen in British Columbia. Likewise, Gustafsson and Eriksson (1995) reported total cation concentrations per gram dry weight for European Aspen in Sweden, but also reported soluble cation concentrations (per gram dry weight) following extraction with SrCl2. We chose to report bark chemistry as soluble nutrient concentrations in water, but on a per unit surface area basis rather than gram dry weight basis in order to reflect the environmental conditions experienced by epiphyte communities upon saturation of bark. Our intention was to approximate the nutrient concentration of recently saturated bark in order to understand variation in nutrients that are leached across the surface of the bark during precipitation events and made available to epiphytes, rather than the nutrient content of the bark itself, which may not all be leachable and certainly not accessible to epiphytes. A reanalysis of our data on bark leachate on a per gram dry weight basis indicated average (± SD) Ca2+, Mg2+, and K+ concentrations to be 83 ± 55, 19 ± 13, and 376 ± 249 μmol * g dry weight-1 of bark, respectively, with Ca2+ and Mg2+ concentrations exhibiting positive correlations (P < 0.001, R2 < 0.17) and K+ concentrations exhibiting no correlation (P = 0.14, R2 = 0.02) with tree diameter. This result adds to inconsistent findings in the literature. Bates (1992) also reported increased Mg bark concentrations (per gram dry weight) with increasing diameter of European Ash, but no relationships for other cations. Gustafsson and Eriksson (1995) reported no correlations between tree diameter and extractable bark cation concentrations Northeastern Naturalist 228 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 for European Aspen. However, Boudreault et al. (2008) reported negative correlations between Quaking Aspen diameter and total bark Ca and Mg concentration on a gram dry weight basis, and Olsson (1978, as cited by Gustafsson and Eriksson 1995) determined that most bark cation concentrations were negatively correlated with age and diameter for Betula pendula Roth (Silver Birch), Picea abies (L.) H. Karst (Norway Spruce), and Pinus sylvestris L. (Scots Pine). With the exception of our study for Ca2+ and Mg2+, and Bates (1992) for Ca, the literature suggests bark cation concentrations are independent of, or decline with increasing tree diameter. This suggests that bark may not typically become more concentrated with cations through time and, in fact, may lose cations, perhaps through leaching by stemflow over extended periods. Importantly, however, our data reveal that even if bark cation concentrations on a dry weight basis remain constant or decline with increasing tree diameter, if bark thickens sufficiently, then cation concentrations of leachate per unit surface area of the bark can still increase as trees age and grow, thereby yielding epiphyte substrate that becomes enriched in certain elemental nutrients. Our method most likely over-estimated nutrient leaching since both outside and inside bark surfaces were exposed during extraction. Elevated rates of nutrient leaching from bark have been reported when samples are not sealed on their inside surfaces (Gustafsson and Eriksson 1995). While differences in methodology limit comparison of our data with that study, our conclusions regarding trends with bark thickness remain valid. Further, while we likely overestimated the amount of nutrients that would be leached solely from the outside bark surface; our chosen methodology to soak 12.6-cm2 bark disks, ranging from 1.5 g to 11.6 g, in 50 g of water may have resulted in solute concentrations that were more dilute than what epiphytes experience immediately upon saturation of the bark during precipitation events. To properly estimate nutrient concentrations experienced by epiphytes at the time of bark saturation, the minute volume of water forming the initial surface film on bark must be analyzed. However, the mesophytic species considered in this study (N. pennata, L. brachypus, P. platyphylla, A. rugelii, and B. laetum) grow as thick wefts and mats. Therefore, the bulk of their photosynthetic tissue is more likely to be exposed to stemflow than to the initial surface film of bark leachate. The averages and variation we measured for solute concentrations in our 102 Sugar Maple bark extracts were similar to the stemflow concentrations of 15 Sugar Maples during the growing season (Fig. 4). Therefore, the concentrations of bark leachate we measured with our methods appear to approach the conditions that epiphytes experience naturally. In this study, cation concentrations in bark leachate increased in relation to tree diameter, but inorganic and organic forms of nitrogen were very low and frequently undetectable, and these did not vary with tree diameter. Of the nitrogen fractions we determined, organic nitrogen was most abundant in the bark extracts. Further, stemflow became enriched with cations, but not with inorganic nitrogen (organic N was not included in stemflow analyses). These findings are consistent with other studies reporting stemflow enrichment by cation nutrients (Andersson 1991, Eaton et al. 1973, Neary and Gizyn 1994, Voigt 1960, Zhang 1989), but not Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 229 by inorganic nitrogen (Neary and Gizyn 1994, Zhang 1989). Neither this nor prior studies measured organic nitrogen in stemflow, but our leachate data suggest that bark may serve as a source of organic nitrogen to stemflow that surpasses the supply of inorganic nitrogen. Although nitrogen nutrition of bryophytes has not been widely studied, Hylocomium splendens has been shown to assimilate both organic (glycine) and inorganic (NH4 + and NO3 -) nitrogen (Forsum et al. 2006). That study showed that 64% of experimentally applied NH4 + was assimilated by H. splendens¸ while 39% of the glycine and 34% of the NO3 - were assimilated. In our study, although DON concentrations in bark leachate were on average 6 times greater than those of NH4 +, it was the less abundant NH4 + that exhibited the few positive correlations with epiphyte cover. Nutrient requirements for the mesophytic and calciphilic species highlighted in this study are not known. However, Basile (1975) recommended that bryophyte culture media be prepared to 250–4300 μmol Ca2+, 700–2200 μmol K+, 200–3000 μmol Mg2+, 2500–7600 μmol NH4 +, and 1000–8500 μmol NO3 -. Further, Bates (1994) suggested that high-nutrient doses of ~5000 μmol L-1 K+, PO4 3-, NH4 +, and NO3 - are sufficient to generate measurable growth responses in bryophytes without inducing severe osmotic imbalances. The average K+ concentration (~900 μmol L-1) of our bark leachate fell within only the lower range of Basile’s recommended value for K+; the average Ca2+ concentration (~200 μmol L-1) approached, but did not meet the minimum recommended value; and the average Mg2+ concentration (~80 μmol L-1) fell short of the recommended value. The average NO3 - and NH4 + concentrations in our bark extract (0.4 μmol L-1 and 5 μmol L-1) were 3–4 orders of magnitude below the recommended concentrations. Even though the concentration of DON was greater than the inorganic fractions in our bark extract, the average molar DON concentration (~31 μmol L-1) was still 2 orders of magnitude less than the recommended inorganic N concentrations. However, bark from the large-diameter Sugar Maples yielded leachate that fell within or exceeded (in the case of K+) Basile’s recommended minimum cation concentrations. Still, maximum nitrogen concentrations were far below recommended levels. Thus, only the larger-diameter Sugar Maples in this study yielded bark leachate that approached recommended concentrations of elemental nutrients for culturing bryophytes. Implications for epiphytic bryophyte communities This study demonstrated that moisture and cation nutrient availability per unit surface area of bark increased with diameter of Sugar Maple trees, and that the cover of several mesophytic and calciphilic bryophyte species increased in relation to tree diameter and 1 or more of the bark moisture or nutrient variables, suggesting that there may be meaningful gradients in time to which individual epiphytic bryophyte species respond. McCune (1993) proposed the similar gradient hypothesis, which predicts that epiphytes respond similarly to 3 different spatial and temporal moisture gradients: vertically on trees within a stand (moisture gradient 1); spatially, between stands of differing moisture regime (moisture gradient 2); and temporally, as a stand develops into the understory reinitiation and old-growth Northeastern Naturalist 230 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 stages (sensu Oliver and Larson 1996) when greater amounts of precipitation can penetrate through forest canopy gaps (moisture gradient 3). Our findings suggest that McCune’s temporal moisture gradient 3 can be explained in other terms. As bark of certain tree species such as Sugar Maple thickens with maturity, it retains more moisture per unit surface-area and dries more slowly. If that difference in moisture availability held within bark is physiologically significant to bryophytes, then thickened, aged bark would provide favorable conditions for succession to mesophytic epiphyte communities. Further, we propose that the similar gradient hypothesis can be conceptually expanded to also include nutrient gradients. Bates (1992), and Gustafsson and Eriksson (1995) demonstrated linkages between soil chemistry, bark chemistry, and epiphyte community composition, thereby providing evidence for an analogous, stand-level, nutrient gradient 2. Our data suggest an analogous temporal nutrient gradient 3 reflecting nutrient enrichment of epiphyte substrate as phorophyte bark thickens with size and age, thereby providing favorable conditions for establishment of calciphilic epiphytes. Finally, if nutrient availability increases through time with aging bark, then a vertical nutrient-gradient may also establish on some trees, reflecting differences in bark conditions across the vertical profile of a tree (nutrient gradient 1). Our data support the hypothesis that temporal moisture and nutrient-resource gradients exist on Sugar Maples that may account for observed succession of epiphyte communities dominated by calciphilic and mesophytic bryophytes on large-diameter Sugar Maple trees. However, alternative hypotheses exist to explain the importance of large trees for maintaining epiphyte diversity. Some researchers (e.g., Boudreault et al. 2008, Hoffman and Boe 1977; but see Mezaka et al. 2008) have concluded that increasing bark roughness with size and age may be more important than bark thickness and moisture-holding capacity for the establishment of mesophytic epiphytes. Bark fissures may provide sheltered microenvironments that are critical for establishment of early protonemata, which may be susceptible to desiccation. Therefore, some epiphytes may be limited to large trees due to their habitat specificity as young gametophytes (i.e., a regeneration niche, sensu Grubb 1977; see also Alpert 1988, Li and Vitt 1995, Slack 1997), rather than ongoing conditions required for established, mature gametophytes. Results of an earlier field-transplant study (McGee and Kimmerer 2004) are consistent with this alternative hypothesis. That study determined that 6-y growth of A. rugelii, P. platyphylla, N. pennata, and L. brachypus mature gametophyte transplants were independent of tree diameter. Therefore, any additional moisture or nutrient availability associated with thick-barked Sugar Maples did not appear to benefit the growth of these transplants, although it is still possible that greater moisture and nutrient availability could improve germination of spores and establishment of protonemata (Armentano and Caponetti 1972, Forman 1964). Some researchers (Löbel et al 2005, Rose 1992, Snäll et al. 2003) proposed that greater epiphyte species richness and unique community composition on large, old trees is a function of colonization probabilities for dispersal-limited organisms. Large, old trees possess greater Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 231 surface areas that are exposed for longer periods of time to intercept rare or poorly dispersed diaspores. It is quite reasonable to expect epiphyte assemblages to reflect interactions between metapopulation dynamics and complex resource-gradients. Conclusions This study demonstrated that the abundance of several mesophytic and calciphilic bryophytes was correlated with increasing Sugar Maple diameter, which in turn covaried with changes in several chemical and physical bark characteristics. In particular, bark-moisture availability and bark leachate cation-nutrient concentrations (per unit surface-area of bark) increased with Sugar Maple diameter, while bark drying rates decreased with diameter. Inorganic and organic nitrogen concentrations were low in bark leachate and did not vary with tree diameter. Stemflow was enriched in dissolved cations, but not inorganic nitrogen, relative to precipitation and throughfall. Only bark samples extracted from large-diameter Sugar Maples produced leachate with cation nutrient concentrations approaching levels required for healthy bryophyte growth in culture. Therefore, these results support the hypothesis that temporal gradients in substrate moisture and nutrient availability on phorophyte hosts result in a succession of epiphytic bryophytes eventually dominated by calciphilic and mesophytic species, and may explain the greater abundance of these species in old-growth northern hardwood forests. However, this correlative study was not able to disentangle, among the several covarying factors, the actual causal mechanisms that might account for observed epiphyte community succession in relation to Sugar Maple diameter. Further, other hypotheses related to colonization probabilities and dispersal limitations of these epiphytes may also explain their observed distributions. The controlled laboratory- and field-based experiments necessary to tease apart these factors will require sowing of sexual and vegetative propagules, and transplants of mature gametophytes, coupled with the technical ability to measure nutrient concentrations and vapor pressure gradients at the microscales that are relevant to these plants. Acknowledgments Funding for this project was provided by the McIntire-Stennis Cooperative Forestry Research Program. Field assistance was provided by E.R. McGee, K. Quinlan, R. Sage, C. Demers, and S. Denniston. We thank 3 anonymous reviewers for their helpful criticisms on a prior draft of this manuscript. Literature Cited Alpert, P. 1988. Survival of a desiccation-tolerant moss, Grimmia laevigata, beyond its observed microdistributional limits. Journal of Bryology 15:219–228. Ameel, J.J., R.P. Axler, and C.J. Owen. 1993. Persulfate digestion for determination of total nitrogen and phosphorus in low-nutrient waters. American Environmental Laboratory 10:1–11. Andersson, T. 1991. Influence of stemflow and throughfall from Common Oak (Quercus robur) on soil chemistry and vegetation patterns. Canadian Journal of Forest Research 21:917–924. Northeastern Naturalist 232 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 Armentano, T.V., and J.D. Caponetti. 1972. The effects of pH on the growth of protonemata of Teraplodon mnioides and Funaria hygrometrica. The Bryologist 75:147–153. Aude, E., and R.S. Poulsen. 2000. Influence of management on the species composition of epiphytic cryptogams in Danish Fagus forest. Applied Vegetation Science 3:81–88. Barkman, J.J. 1958. Phytosociology and Ecology of Cryptogamic Epiphytes. Koninklyke Van Gorcum and Company, NV Assen, The Netherlands. 628 pp. Basile, D.V. 1975. A comparison of some macronutrient media used to culture bryophytes. The Bryologist 78:403–413. Bates, J.W. 1992. Influence of chemical and physical factors on Quercus and Fraxinus epiphytes at Loch Sunart, western Scotland: A multivariate analysis. Journal of Ecology 80:163–179. Bates, J.W. 1994. Responses of the mosses Brachythecium rutabulum and Pseudoscleropodium purum to a mineral nutrient pulse. Functional Ecology 8:686–693. Billings, W.D., and W.B. Drew. 1938. Bark factors affecting the distribution of corticolous bryophyte communities. American Midland Naturalist 20:302–330. Boudreault, C., D.S. Coxson, E. Vincent, Y Bergeron, and Y. Marsh. 2008. Variation in epiphytic lichen and bryophyte composition and diversity along a gradient of productivity in Populus tremuloides stands of northeastern British Columbia, Canada. Ecoscience 15:101–112. Braun, E.L. 1950. Deciduous Forests of Eastern North America. Hafner Publishing Company, New York, NY. 596 pp. Brown, D.H. 1982. Mineral nutrients. Pp. 383–444, In A.J.E. Smith (Ed.). Bryophyte Ecology. Chapman and Hall, New York, NY. 512 pp. Cain, S.A., and A.J. Sharp. 1938. Bryophytic unions of certain forest types of the Great Smoky Mountains. American Midland Naturalist 20:249–301. Cleavitt, N.I., A.C. Dibble, and D.A. Werier. 2009. Influence of tree composition upon epiphytic macrolichens and bryophytes in old forests of Acadia National Park, Maine. The Bryologist 112:467–487. Crum, H. 1991. Liverworts and Hornworts of Southern Michigan. University of Michigan Herbarium, Ann Arbor, MI. 233 pp. Culberson, W.L. 1955. The corticolous communities of lichens and bryophytes in the upland forests of northern Wisconsin. Ecological Monographs 25:215–231. Eaton, J.S., G.E. Likens, and F.H. Bormann. 1973. Throughfall and stemflow chemistry in a northern hardwood forest. Journal of Ecology 61:495–508 Everhart, S.E., J.S. Ely, and H.W. Keller. 2009. Evaluation of tree-canopy epiphytes and bark characteristics associated with the presence of corticolous myxomycetes. Botany 87:509–517. Flora of North America Editorial Committee (Eds.; FNA). 2007. Flora of North America North of Mexico. Volume 27: Bryophytes. Oxford University Press, New York, NY, and Oxford, UK. 734 pp. FNA. 2014. Flora of North America North of Mexico. Volume 28: Bryophytes. Oxford University Press, New York, NY, and Oxford, UK. 736 pp. Forman, R.T.T. 1964. Growth under controlled conditions to explain the hierarchical distributions of a moss, Tetraphis pellucida. Ecological Monographs 34:1–25. Forsum, A., L. Dahlman, T. Nasholm, and A. Nordin. 2006. Nitrogen utilization by Hylocomnium splendens in a boreal forest-fertilization experiment. Functional Ecology 20:421–426. Freund, R.J., and R.C. Littell. 1991. SAS System for regression. SAS Institute, Inc., Cary, NC. 210 pp. Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 233 Fritz, Ö., M. Niklasson, and M. Churski. 2009. Tree age is a key factor for the conservation of epiphytic lichens and bryophytes in beech forests. Applied Vegetation Science 12:93–106. Gimingham, C.H., and E.M. Birse. 1957. Ecological studies on growth-form in bryophytes. I. Correlations between growth form and habitat. Journal of Ecology 45:533–545. Grubb, P.J. 1977. The maintenance of species richness in plant communities: The importance of the regeneration niche. Biological Reviews 52:107–145. Gupta, R.K. 1977. A study of photosynthesis and leakage of solutes in relation to the desiccation effect in bryophytes. Canadian Journal of Botany 55:1186–1194. Gustafsson, L., and I. Eriksson. 1995. Factors of importance for the epiphytic vegetation of Aspen, Populus tremula, with special emphasis on bark chemistry and soil chemistry. Journal of Applied Ecology 32:412–424. Hale, M.E. 1955. Phytosociology of corticolous cryptogams in the upland forests of southern Wisconsin. Ecology 36:45–63. Halpern, C.B., M. Dovčiak, L.S. Urgenson, and S.A. Evans. 2014. Substrates mediate responses of forest bryophytes to a gradient in overstory retention. Canadian Journal of Forest Research 44:855–866. Haughian, S.R., and K.A. Frego. 2017a. Does CWD mediate microclimate for epixylic vegetation in boreal forest understories? A test of the moisture-capacitor hypothesis. Forest Ecology and Management 389:341–351. Haughian, S.R., and K.A. Frego. 2017b. Log moisture capacity does not predict epixylic bryophyte growth under thinned and unthinned forest canopies. Functional Ecology 31:1540–1549. Hazell, P., O. Kellner, H. Rydin, and L. Gustafsson. 1998. Presence and abundance of four bryophytes in relation to density of Aspen (Populus tremula) and other stand characteristics. Forest Ecology and Management 107:147–158. Heimburger, C.C. 1934. Forest-Type Studies in the Adirondack Region. Cornell University Agriculture Research Station Memoir 165, Ithaca, NY.. Hoffman, G.R., and A.A. Boe. 1977. Ecological study of epiphytic cryptogams on Populus deltoides in northeastern South Dakota and adjacent Minnesota. The Bryologist 80:32–47. Hosokawa, T., and N. Odani. 1957. The daily compensation period and vertical ranges of epiphytes in a beech forest. Journal of Ecology 45:901–905. Hosokawa, T., N. Odani, and H. Tagawa. 1964. Causality of the distribution of corticolous species in forests with special reference to the physio-ecological approach. The Bryologist 67:396–411. James, P.T., B.J. Crandall-Stotler, and R.E. Stotler. 1998. Morphological and genetic variation in Porella platyphylla and P. platyphylloidea and their systematic implications. The Bryologist 101(1):1–19. Király, I., and P. Ódor. 2010. The effect of stand structure and tree species composition on epiphytic bryophytes in mixed deciduous–coniferous forests of Western Hungary. Biological Conservation 143:2063–2069. Levia, D.F., and S.R. Herwitz. 2005. Interspecific variation of bark-water storage capacity of three deciduous tree species in relation to stemflow yield and solute flux to forest soils. Catena 64:117–137. Li, Y., and D.H. Vitt. 1995. The dynamics of moss establishment: Temporal responses to a moisture gradient. Journal of Bryology 18:677–687. Littell, R.C., R.J. Freund, and P.C. Spector. 1991. SAS System for Linear Models. SAS Institute, Inc. Cary, NC. Northeastern Naturalist 234 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 Vol. 26, No. 1 Lobel, S., T. Snall, and H. Rydin. 2005. Species richness patterns and metapopulation processes: Evidence from epiphyte communities in boreo-nemoral forests. Ecography 29:169–182. 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 M.J. Mefford. 2011. PC-ORD. Multivariate analysis of ecological data. Version 6. MjM Software, Gleneden Beach, OR. McGee, G.G., and R.W. Kimmerer. 2002. Forest age and management effects on epiphytic bryophyte communities in Adirondack northern hardwood forests, New York, USA. Canadian Journal of Forest Research 32:1562–1576. McGee, G.G., and R.W. Kimmerer. 2004. Size of Acer saccharum hosts does not influence growth of mature bryophyte gametophytes in Adirondack northern hardwood forests. The Bryologist 107:302–311. McGee, G.G., D.J. Leopold, and R.D. Nyland. 1999. Structural characteristics of oldgrowth, maturing, and partially cut northern hardwood forests: Implications for ecosystem management. Ecological Applications 9:1316–1329. Mežaka, A., G. Brūmelis, and A. Piterāns. 2008. The distribution of epiphytic bryophytes and lichen species in relation to phorophyte characters in Latvian natural old-growth broad-leaved forests. Folia Cryptogamica Estonica 44:89–99. Mežaka, A., G. Brūmelis, and A. Piterāns. 2012. Tree- and stand-scale factors affecting richness and composition of epiphytic bryophytes and lichens in deciduous woodland key habitats. Biodiversity and Conservation 21:3221–3241. Neary, A.J., and W.I. Gizyn. 1994. Throughfall and stemflow chemistry under deciduous and coniferous forest canopies in south-central Ontario. Canadian Journal of Forest Research 24:1089–1100. Ódor, P., I. Király, F. Tinya, F. Bortignon, and J. Nascimbene. 2014. Patterns and drivers of species composition of epiphytic bryophytes and lichens in managed temperate forests. Forest Ecology and Management 321:42–51. Oliver, C.D., and B.C. Larson. 1996. Forest Stand Dynamics. John Wiley and Sons, Inc. New York, NY. 544 pp. Olsson, M. 1978. Properties and decomposition of bark. Reports in Forest Ecology and Forest Soils. Swedish University of Agricultural Sciences. Uppsala, Sweden. Phillips, E.A. 1951. The associations of bark-inhabiting bryophytes in Michigan. Ecological Monographs 21:301–316. Proctor, M.C.F. 1982. Physiological ecology: Water relations, light and temperature responses, carbon balance. Pp. 333–381, In A.J.E. Smith (Ed.). Bryophyte Ecology. Chapman and Hall, New York, NY. 512 pp. Proctor, M.C.F., M.J. Oliver, A.J. Wood, P. Alpert, L.R. Stark, N.L. Cleavitt, and B.D. Mishler. 2007. Desiccation-tolerance in bryophytes: A review. The Bryologist 110:595– 621. Richardson, D.H.S. 1981. The Biology of Mosses. John Wiley and Sons, New York, NY. 232 pp. Rose, F. 1992. Temperate forest management: Its effect on bryophyte and lichen floras and habitats. Pp. 211–233, In J.W. Bates and A.M. Farmer (Eds.) Bryophytes and Lichens in a Changing Environment. Clarendon Press, Oxford, UK. 416 pp. Slack, N.G. 1976. Host specificity of bryophytic epiphytes in eastern North America. Journal of the Hattori Botanical Laboratory 41:107–132. Slack, N.G. 1997. Niche theory and practice: Bryophyte studies. Advances in Bryology 6:169–204. Northeastern Naturalist Vol. 26, No. 1 G.G. McGee, M.E. Cardon, and D.H. Kiernan 2019 235 Small, H., T.S. Stevens, and W.C. Bauermann. 1975. Novel ion-detection chromatographic method using conductimetric detection. Analytical Chemistry 47:1801. Snäll T., P.J. Ribeiro, and H. Rydin. 2003. Spatial occurrence and colonizations in patchtracking metapopulations: Local conditions versus dispersal. Oikos 103:566–578. Snäll, T., A. Hagström, J. Rudolphi, and H. Rydin. 2004. Distribution pattern of the epiphyte Neckera pennata on tree spatial scales: Importance of past landscape structure, connectivity, and local condition. Ecography 27:757–766. Somers, R.C., Jr. 1986. Soil classification, genesis, morphology, and variability of soils found within the central Adirondack region of New York. Ph.D. Dissertation. State University of New York, College of Environmental Science and Forestry, Syracuse, NY. Studlar, S.M. 1982a. Host specificity of epiphytic bryophytes near Mountain Lake, Virginia. The Bryologist 85:37–50. Studlar, S.M. 1982b. Succession of epiphytic bryophytes near Mountain Lake, Virginia. The Bryologist 85:51–63. Szövényi, P., Z.S. Hock, and Z. Tóth. 2004. Phorophyte preferences of epiphytic bryophytes in a stream valley in the Carpathian Basin. Journal of Bryology 26:137–146. US Environmental Protection Administration (USEPA). 1983. Methods for chemical analysis of waters and wastes. EPA-600/4-79-020. Environmental Monitoring and Support Laboratory. Cincinnati, OH. Voigt, G.K. 1960. Alteration of the composition of rainwater by trees. American Midland Naturalist 63:321–326. Zhang, Y. 1989. Throughfall and stemflow chemistry in a northern hardwood forest in the Adirondack Region of New York. M.Sc. Thesis. State University of New York, College of Environmental Science and Forestry, Syracuse, NY.