2013 NORTHEASTERN NATURALIST 20(Monograph 10):1–60 The Invasive Ailanthus altissima in Pennsylvania: A Case Study Elucidating Species Introduction, Migration, Invasion, and Growth Patterns in the Northeastern US Matthew T. Kasson1,2,*, Matthew D. Davis1, and Donald D. Davis1 Abstract - Ailanthus altissima (Tree of Heaven), an invasive tree species native to China and East Asia, was first introduced into the US ca. 1784 by William Hamilton at his Philadelphia, PA estate. However, the means and temporal progression of spread from this and other early points of introduction are not clear. This species now occurs in >40 US states, primarily as an urban and roadside weed. The Northeast supports the highest densities of Ailanthus within the US, mainly in transportation corridors and urban areas, where it has become the dominant tree species. A recent, widespread increase in Ailanthus incidence in eastern hardwood forests, not unlike prior invasions along railways and roadsides, suggests that current conditions favor invasion in natural environments. To help elucidate the life history of Ailanthus in Pennsylvania and the northeastern US, as well as answer fundamental biological questions concerning this species, we conducted dendrochronological (tree-ring) studies and floristic surveys beginning in 2010. Although we studied population dynamics, age structure, and tree-ring characteristics of Ailanthus primarily in Pennsylvania, we supplemented our studies using trees from adjacent northeastern states. Floristic studies, conducted between 2010 and 2011, revealed Ailanthus to be present in 60 of 67 Pennsylvania counties, including nine previously unreported counties. Tree-ring studies of these trees, as well as of trees from adjacent states, indicated Ailanthus was a suitable species for tree-ring analyses, although false rings were observed in >20% of trees examined. Our results also revealed Ailanthus to be longer-lived than previously reported, reaching ages >100 years, at which time female trees still produced viable seed. However, extant Ailanthus did not exceed 120 years in age, and therefore could not be directly linked with any documented late 18th-century or early 19th-century plantings. Nevertheless, continuous colonization in southeastern Pennsylvania, spanning two centuries, by overlapping generations of Ailanthus was supported by observations at Bartram’s garden in Philadelphia. Regression analysis revealed that tree diameter was a significant predictor (R2 = 0.80) of Ailanthus age and could be used to estimate ages of historic trees whose diameters were reported in early literature, which revealed establishment of Ailanthus in six Pennsylvania counties 70–118 years earlier than previously reported. Tree-ring studies of extant trees revealed that Ailanthus had gone unreported in some counties for 50 years, during which subsequent invasions occurred. Lastly, we report that widespread invasion of forests by Ailanthus is a relatively recent phenomenon in Pennsylvania, with most invasions occurring after 1965, following major forest disturbances such as salvage logging in the aftermath of widespread insect defoliator outbreaks, especially Lymantria dispar (Gypsy Moth) defoliations. These findings are consistent with previously reported lag times between initial colonization and the onset of rapid range expansion for other invasive plant species. Our results emphasize the need 1The Pennsylvania State University, Department of Plant Pathology, 401 Buckhout Lab, University Park, PA 16802. 2Current address - Virginia Tech, Department of Plant Pathology, Physiology, and Weed Science, 115 Price Hall, Blacksburg, VA 24061. *Corresponding author - mkasson@vt.edu. 2 Northeastern Naturalist Vol. 20, Monograph 10 for more thorough inventories of Ailanthus within forests prior to harvesting. Such inventories will afford forest managers opportunities to preemptively eradicate Ailanthus and mitigate future invasions by this species. Introduction Early introductions and observations of Ailanthus in the Northeast Ailanthus altissima (Mill.) Swing. (formerly A. glandulosa Desf.) (Tree of Heaven), hereafter called Ailanthus, was first introduced into the US from its native China via England in 1784 or 1785 by William Hamilton at his country estate “The Woodlands” (currently the Woodlands Cemetery) in Philadelphia, PA (Browne 1846, Carr 1861, Dwight 1845). A second introduction into the northeastern US occurred in 1804 in Portsmouth, RI, directly from China (Browne 1846, Prince 1827). Despite these early introductions into the northeastern US, it wasn’t until several decades later that Ailanthus became widely available in the horticultural trade: first and foremost by Prince Nursery (Flushing, Long Island, NY) in 1820, followed by Bartram’s Garden (Philadelphia, PA) in c. 1828 (Browne 1846, Carr 1828). Widespread adoption of Ailanthus by nurserymen and gardeners after 1820 was likely due to the presence of seed-bearing Ailanthus, which allowed trade in seed and seedlings rather than cuttings. Indeed, the earliest evidence of seed production by Ailanthus in the US is found in an undated herbarium specimen from the Philadelphia Herbarium at the Academy of Natural Sciences, Philadelphia (PH) collected by Solomon W. Conrad in Philadelphia between 1815 and 1831 (Fig. 1a, cover). Given the rarity of Ailanthus during this time in Philadelphia, the specimen likely came from Bartram’s garden, the only confirmed site of Ailanthus cultivation in Philadelphia prior to 1827 in addition to the Woodlands. During this same time period, Conrad had collected other plant species in Bartram’s Garden that are also included in collections at the Philadelphia Herbarium (Rabeler 1985). Procurement of Ailanthus at Bartram’s garden for out-planting at the Philadelphia Cemetery in 1827 support the likelihood of seed propagation at this time (Jones 1835). A second specimen from Philadelphia, dated 1841, accessioned in the New York Botanical Garden Herbarium (NY Accession No. 785400), represents the earliest, accurately dated seed-bearing Ailanthus in the US (Fig. 1b). Other early seed-bearing specimens from surrounding areas include Batsto Village, NJ (1863) and Brooklyn, NY (1866) (NY Accession No. 785391 and Brooklyn Botanical Garden Herbarium [BKL] Accession No. 66572, respectively). Based on these specimens, seed-bearing Ailanthus were not likely present before 1820, since Ailanthus seed had not been previously reported or found in any extant US herbaria collections. Moreover, Ailanthus trees <10 years of age can bear seed, which would have given this species ample opportunity to sexually reproduce, had both gynoecious (seed-producing) and androecious (pollen-producing) plants been present in the areas of early introduction (Dwight 1845, Miller 1835). The earliest reports of seed-bearing Ailanthus in Pennsylvania (PA) were first published in 1846 (Browne 1846) and with specific reference to Bartram’s garden 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 3 in 1855 (Barry and Smith 1855). By 1868, Ailanthus’ ability to spread from seed in PA, and presumably elsewhere, was widely acknowledged (Gray 1868b). One explanation for the absence of sexual reproduction during the first decades following introduction comes from Monson (1843), who explained that Ailanthus planted in cities during the early 19th century were mainly cuttings from a single “variety” that did not produce seed. In Philadelphia, prior to seed cultivation by Bartram and other nurserymen, Ailanthus were propagated exclusively by cuttings from Hamilton’s original staminate trees (Browne 1846, Dwight 1845, JBA 1907). Herbarium specimens from England’s earliest points of introduction (ca. 1751), Chelsea Botanic Garden,and Philip C. Webb’s private exotic gardens at Busbridge in Surrey near London, support the pervasiveness of staminate Ailanthus during the time period when Figure 1. A) Earliest known Ailanthus altissima herbarium specimen in the US containing seed. Collected by S.W. Conrad, Philadelphia, Philadelphia County, PA, between 1815–1831 (Philadelphia Herbarium at the Academy of Natural Sciences, Accession No. 01097312). Notes: Decand. trigynia (polygam), S.W. Conrad. Although undated, this specimen includes Conrad’s pre-printed commercial labels that he began using after 1815 (Wilson 2011). Conrad used Linnaean classification to identify the plant. Conrad’s classification was later supported by a formal botanical description of Ailanthus (Gray 1868a). Printed with permission of the Academy of Natural Sciences, Philadelphia B) Earliest accurately dated US Ailanthus altissima herbarium specimen containing seed, Collected by J.P. James, Philadelphia, Philadelphia County, PA, August 1841 (New York Botanical Garden Herbarium, Accession No. 785400). Notes: Ex Herb. J.M. Coulter, Incorporated Herbarium: Herbarium of Wabash College. Printed with permission of the New York Botanical Garden Herbarium. 4 Northeastern Naturalist Vol. 20, Monograph 10 Hamilton’s Ailanthus was acquired from England (Loudon 1844, Miller 1835) (Natural History Museum, London Accession No. BM001017992). Despite the ease of propagation, Hamilton apparently disseminated few of his Ailanthus plants. Hamilton jealously guarded these rare exotic plants, since he went to considerable risk and trouble to procure and import them to the US (Browne 1846, Downing 1844). In a letter to his secretary Benjamin H. Smith (dated 12 June 1790), Hamilton wrote, “It would have been an agreeable circums tance to me to have heard the large sumachs [sic, Ailanthus] and Lombardy poplars … have not been neglected. After the immense pains I took in removing the exotics to the north front of the house, by way of experiment, you will naturally suppose me anxious to know [their] success … and the effect as to [their] appearance in approach …” (Smith 1905b). In another letter, Hamilton emphasized that plants “… should be kept in the Hot House under lock and key … [and] nobody … should get sight of them …” (Smith 1905a). Consequently, only Hamilton’s horticulturally minded contemporaries including Humphrey Marshall and William Bartram ever received Ailanthus cuttings from Hamilton, and only after sharing plants from their own collections (Browne 1846, Darlington 1849). In 1809, Bartram received an Ailanthus sucker sprout from Hamilton. By 1877, the sprout had grown to a sizable 90.7 cm in diameter at breast height Figure 2. Historical measurements of Ailanthus trees found throughout PA from time of introduction through 2010. A value of 0.0 was given for trees with a known planting date and for individuals whose measurements were excluded from a specific reference but contained, in many cases, a qualitative description. Reference line at 107 cm denotes the maximum diameter previously reported for Ailanthus (Miller 1990). Some individual observations represent more than one tree. Other denotes both historic and non-historic locations and includes observations across ten PA counties (See Appendix 2 for complete list of references). 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 5 (DBH: diameter at 1.4 m; Fig. 2; Browne 1846, Sargent 1878). Observations in Marshall’s arboretum ca. 1893 revealed “… a very large Ailanthus, which must have been one of the first specimens planted in America [and] seems to date from the time of Marshall,” ca. early 19th century (Sargent 1893). Although the diameter of Marshall’s Ailanthus in 1888 was not reported, comparisons to other old trees measured in Marshall’s arboretum in 1880 allows a conservative estimation of 60–75 cm for Marshall’s Ailanthus (Fig. 2; Sargent 1880). Further dissemination of Hamilton’s Ailanthus plants, prior to distribution in the horticultural trade, may have resulted from an 1809 exotic plant sale from Hamilton’s extensive collection, or as Ailanthus cuttings from Bartram’s or Marshall’s established specimens (Belden 1958, Darlington 1853, Dwight 1845, Hamilton 1809). Reports of comparably sized Ailanthus trees at the Woodlands through the 19th and 20th century were not found, suggesting that Hamilton’s own Ailanthus trees were either short–lived and/or cut back purposely to provide a desired aesthetic in his meticulously manicured English garden (Long 1991, Oldschool 1809). A third possibility building on the latter hypothesis, and supported by several accounts including Hamilton’s own letters, was that Ailanthus trees among other exotics were never out-planted on the grounds of The Woodlands but instead “… were arranged tastefully in large oval grass plats (artificial mounds with varioussized holes for potted plants) in front of the conservatory [to] represent a miniature hill clothed with choice vegetations” (Drayton 1806), after which plants were returned to the greenhouses for the winter (Long 1991, Wunsch 2003). Reports that Bartram and Henry Pratt (Lemon Hill, Fairmont Park, Philadelphia) received Ailanthus suckers from Hamilton at the time of the documented plant sale, support the latter hypothesis that some of Hamilton’s Ailanthus trees were never out-planted, but rather were likely distributed to Hamilton’s fellow botanists (Browne 1846, Dwight 1845). Regardless of uncertainty involving the spatial arrangement and persistence of first-generation Ailanthus stems at the Woodlands, Harshberger (1921) observed “… descendants of the first Ailanthus tree planted in America …” at the Woodlands in c. 1920. Whether the Ailanthus was established from winddisseminated seed from off-site, or as sprouts of surviving root systems from first generation plantings, is unknown. Unfortunately, Harshberger reported no additional information regarding Ailanthus tree location or size, nor are there any records documenting the original planting sites for Ailanthus at the Woodlands. Commercialization of Ailanthus Shortly after Ailanthus became widely available in the horticultural trade, demand for this species surged as evidenced by numerous newspaper advertisements throughout New England and the Mid-Atlantic region, which were absent throughout the 1820s (Anonymous 1833, Manning 1830, Minor 1835, Pavilion 1830, Peirce 1833, Sinclair 1835, Thornburn 1830). This surge in popularity was attributed to Ailanthus’ noble appearance and tropical-like foliage, hardiness and rapidity of growth, and for its ability to thrive in the poorest soil (Sargent 1888). As testament to the high demand for this species, 6 Northeastern Naturalist Vol. 20, Monograph 10 Bartram’s 1828 Garden catalogue listed Ailanthus at $1.00 each compared to $0.25 each for Populus nigra “Italica” L. (Lombardy Poplar) and $0.50 each for Aesculus hippocastanum L. (Horse Chestnut), two other popular imports (Carr 1828). Only Ginkgo biloba L. (Ginkgo, Maidenhair Tree), another exotic import credited to Hamilton, and a native, yellow-flowered Magnolia acuminata L. (Cucumber Magnolia), demanded a higher price at this time (Carr 1828). More than 15 years later, a report from NY by Dwight (1845) listed Ailanthus at $1.50 each for 2-year-old trees, indicating that demand for this species was still very high into the middle of the 19th century. Although Ailanthus gained popularity across smaller cities and towns in the mid-1800s, its notoriety in large cities where it was first introduced became readily apparent. In 1852, A.J. Downing, prominent writer and editor of The Horticulturist and Journal of Rural Art and Rural Taste, denounced the incurable vices of Ailanthus. Downing (1852) spoke out firmly against further planting of Ailanthus, insisting it “… fills the air with a heavy, sickening odor, it suckers abominably, and thereby over runs, appropriates and reduces to beggary, all the soil of every open piece of ground where it is planted …” Yet despite the problems with Ailanthus in urban centers where it was aggressively planted at the cost of native species, “… country gentlemen, seduced by the oriental beauty of its foliage, continue to plant it in their pleasure grounds” (Downing 1852). Surprisingly, Ailanthus is still available (2013) commercially via various online nurseries for as little as $1.50 each for 1–2 foot tall trees. Intra-state migration, proliferation, and growth of Ailanthus in PA Even with high costs and growing opposition in cities, the incidence of Ailanthus throughout PA increased steadily after 1820 (Appendix 1). Concurrently, the incidence of large Ailanthus trees (>50 cm DBH) was reported for the first time beginning around 1850 (Fig. 2). By the early 20th century, reports indicate that Ailanthus was well distributed throughout the state, suggesting that many of the scattered geographic pockets reported in the mid- to late-1800s of Ailanthus had presumably coalesced (Appendix 1). The incidence of larger Ailanthus trees (50–120 cm DBH) also increased dramatically in southern PA during this time period, with reports of large Ailanthus in Adams, Chester, Cumberland, Dauphin, Franklin, and Philadelphia counties (Fig. 2; Illick and Brouse 1926). Large trees at Bartram’s and Marshall’s gardens were, given their growth patterns, likely second-generation root sprouts of Ailanthus planted earlier at these same sites (Fig. 2, 3B). Large, first-generation Ailanthus trees were reported in Adams, Cumberland, Dauphin, and Franklin counties, further validating species acclimation throughout its new environment (Figs. 2, 3C). Naturalized dense thickets of Ailanthus also were observed near Reading (Berks County), Mont Alto (Franklin County), and north of Harrisburg along the Susquehanna River upstream from Clarks Ferry Bridge (Dauphin County) (Illick and Brouse 1926). Reforestation projects using Ailanthus seed on denuded industrial lands in western PA by the PA Division of Forests and Waters (now PA Bureau of Forestry) likely hastened Ailanthus’ expansion into additional disturbed areas of the state (Illick and Brouse 1926). 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 7 Over the last half-century, published reports indicate a dramatic increase in counties with confirmed Ailanthus (Appendix 1). Yet, during this same time period, the largest Ailanthus trees ever recorded in the state were reported from two counties (Lebanon and Wyoming) where Ailanthus allegedly had established after 1960 (Fig. 2, Appendix 2; Clark 1988, 1991). The apparent inaccuracies in county-specific incidence reports for Ailanthus emphasize the need for further investigation to more firmly resolve spatiotemporal migration patterns. Given the long residency (≈230 years) of this species in the state, it is likely that Ailanthus populations may have occurred at low levels (e.g., individual trees) earlier in many additional counties with first reports after 1960, but went unnoticed or unreported. Invasion of PA forests It is unclear when Ailanthus first became a significant problem on PA forest lands and in rest of the Northeast despite Ailanthus’ presence as an incidental species in many counties and states for more than a century. Currently, Ailanthus Figure 3. Early photographs of Ailanthus trees (white arrows) in PA: A) Krauth House, Lutheran Theological Seminary, Gettysburg ca. 1911–1925; B) Humphrey Marshall’s arboretum, Marshallton c. 1925; C) York Springs ca. 1925. The Krauth House trees were two of eight apparently planted at the newly established Seminary during the early to mid-1830s, but were cut down in April 1950 because of declining health. At Marshall’s garden there was a group of seven large Ailanthus of spout origin ranging from 61–76 cm DBH in ca. 1925. Ailanthus was first introduced here before 1820. The last documented Ailanthus in Marshall’s arboretum was an 86.4-cm-diameter stump in 1958. The Ailanthus tree at York Springs was 91.4 cm DBH in 1925. Photo A printed with permission of Lutheran Theological Seminary. Photos B and C reprinted from Illick and Brouse (1926) with permission of the Pennsylvania Department of Conservation and Natural Resources, Bureau of Forestry. 8 Northeastern Naturalist Vol. 20, Monograph 10 is one of 25 reported non-indigenous trees and shrub species intentionally introduced into the PA for food, fiber, or ornamental purposes that have escaped cultivation and pose serious, long-term threats to native plant diversity (Morse et al. 1995, PA DCNR 2013). Ailanthus’ allelopathic properties against 35 deciduous and 34 coniferous native species (Mergen 1959), prolific seed production (Bory and Clair-Maczulajtys 1980), and persistent root suckering (Hu 1979) make it an exceptional invader, particularly in open areas with disturbed mineral soil, where Ailanthus easily supplants native species (Rhoads and Block 2007). Successful invasion may be further enhanced by the species’ association with endomycorrhizal fungi (Huebner et al. 2007). Ailanthus can also exploit small forest gaps where it relies on vegetative reproduction, surviving in the understory as long as 19 years prior to recruitment (Knapp and Canham 2000, Kowarik 1995, Martin et al. 2010). Because most state and federal forest lands in PA were not established until the early 1900s, Ailanthus was likely present on some lands prior to their acquisition by governmental agencies. In addition, some Ailanthus were probably intentionally introduced onto public lands as a means of reclaiming post-industrial waste lands (Illick and Brouse 1926). Data from USDA Forest Service Forest Inventory and Analysis (FIA) plots recently revealed that Ailanthus stems increased dramatically in PA forests from 76 million in 1989 to ≈135 million in 2004. These findings suggest that conditions since 1989 favor proliferation of Ailanthus in PA forests where this species had been significantly less abundant or previously absent (Alerich 1993, McWilliams et al. 2007). Despite these reported increases in Ailanthus stems from 1989 to 2004, data from southeastern PA (SE PA) did not distinguish Ailanthus in the 1989 inventory where many stands were identified in the 2004 inventory. Thus, comparisons between the two inventories could not be made. Nevertheless, significant increases did occur where Ailanthus was documented in both inventories (Alerich 1993, McWilliams et al. 2007). Eichelberger and Perles (2009) recently evaluated the status of invasive plant species in the Delaware Water Gap National Recreation Area (DEWA) in PA and New Jersey (NJ). These authors listed Ailanthus as a top invasive plant, particularly in shale woodlands, shale scree slopes, and sparsely vegetated cliff areas. Turner et al. (2007) documented recent establishment of Ailanthus within the forest understory at the Gordon Natural Area, Chester County, PA, which has occurred since a previous survey in 1970 (Overlease 1973). Similarly, Espenschied- Reilly and Runkle (2008) reported an increase in incidence and size class of Ailanthus from 1980 to 2002 within the Wright State University woodlot near Dayton, OH. These observations, among others, reveal a disconcerting trend of increasing invasion by Ailanthus into native forests and natural landscapes. Even in the absence of major disturbances, spread of invasive species such as Ailanthus continues to occur in areas adjacent to roadsides (Mortensen et al. 2009). Transportation corridors are ideal for colonization by invasive species due to their open nature, but also due to secondary seed dispersal mediated by vehicular traffic (Aldrich et al. 2010). With continued suburban sprawl and 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 9 infrastructure expansion, not only will these trends continue but will likely intensify as well. Persistence of Ailanthus on the landscape Effective control strategies to mitigate further spread of Ailanthus are limited by our incomplete understanding of biological processes that enhance survival and proliferation of this species. One key to understanding invasion potential is to answer fundamental questions regarding Ailanthus life expectancy and growth patterns. With knowledge of such attributes, we can more accurately reconstruct Ailanthus spatiotemporal migration throughout PA and evaluate factors that favored the establishment of extant populations. Such knowledge can ultimately guide management and control efforts in PA where Ailanthus has long been established, as well as elsewhere in the US where Ailanthus has only recently invaded. Despite previous reports that Ailanthus is short-lived in the northeastern US, ranging from 50–75 years (Jellett 1904, Sargent 1878, Stefferud 1949), dendrochonological (tree-ring) investigations in England (Cutler et al. 1993) and anecdotal evidence in the US (Belden 1958, Sargent 1893, Wentz 1950) suggests that the lifespan of Ailanthus likely exceeds 100 years. In southeastern China, Ailanthus’ country of origin, 70-year-old trees have been reported (Shouxiang et al. 2008). However, no thorough dedrochronological investigation of this species has been conducted in the US to corroborate reports of longer-lived Ailanthus in the US. Nevertheless, large Ailanthus trees throughout the eastern US have been reported that far exceed the previously established maximum 107 cm DBH (Fig. 2; Illick and Brouse 1926), from which some of the earliest age estimations for Ailanthus were inferred. Research objectives The goal of our project was to elucidate where Ailanthus occurs, where long-established individuals and populations persist, and what factors have initiated recent widespread invasion throughout PA forests. We had two main objectives: 1) determine feasibility of using Ailanthus tree-ring chronologies to aid in reconstruction of tree and stand/site histories throughout the Northeast, determine the maximum age of Ailanthus, and explore age/diameter relationships for historical inference and temporal reconstruction; and 2) evaluate county-level incidences of extant Ailanthus populations throughout PA with emphasis along highway corridors, in urban and suburban epicenters where historic populations of Ailanthus are well documented, and within forests where recent Ailanthus invasions have been reported. Using these data along with historic records and accounts, an up-to-date spatiotemporal timeline for Ailanthus range expansion throughout PA from the earliest points of introduction forward to recent invasions of forest lands was generated. Methods Study area The Mid-Atlantic region of the eastern US including Delaware (DE), Maryland (MD), NJ, PA, Virginia (VA), and West Virginia (WV) along with the eastern 10 Northeastern Naturalist Vol. 20, Monograph 10 edge of the Midwest region (Indiana [IN], Ohio [OH]) and the northern edge of the Southeast region (Kentucky [KY], Tennessee [TN]), supports the highest densities of Ailanthus in the US (Invasive Plant Atlas of the US, http://www. invasiveplantatlas.org/). Since Ailanthus was first introduced into PA almost 230 years ago, this region may support numerous large and presumably old Ailanthus trees from which life history patterns can be obtained using tree-ring chronologies. The main tree-ring study herein includes Ailanthus trees from DE, NY, OH, PA, VA, and WV with emphasis on SE PA (Fig. 4). Within SE PA, trees were sampled within, or adjacent to, points of early introduction, including trees from the Woodlands Cemetery where Ailanthus was first introduced in c. 1784, Bartram’s garden where an Ailanthus sucker from Hamilton’s tree was planted in 1809, and from Pratt’s Garden (Lemon Hill Mansion) where Ailanthus was planted in the first decades of the 19th century (Fig. 2, Appendix 2). The Ailanthus study area includes all counties in PA and builds on previously characterized populations (McWilliams et al. 2007, Rhoads and Klein 1993, Wherry et al. 1979). A second tree-ring study was implemented to determine when Ailanthus first invaded state forest lands; this study included Ailanthus of various ages and sizes (Fig. 4). Dendrochronology Tree Core Data. To locate the largest Ailanthus trees throughout the study area, we utilized State Big Tree Registry reports (Hill 1986, Wade 2006), online databases (e.g., http://www.pabigtrees.com/), and referrals from state registry coordinators and measurements crews, state foresters, arboretum staff, arborists, and state and city park invasive species specialists. We also established contacts with grounds managers at private estates throughout SE PA where large Ailanthus had been observed. Letters were mailed to private residences to obtain permission to sample large candidate trees. A minimum of one radial core was extracted from each Ailanthus tree using a 76-cm, 5.15-mm diameter Haglof increment borer (Haglof Inc., Madison, MS). If internal rot was detected during core extraction, a second, and occasionally a third, increment core was taken if permission was granted by the owner. All cores, including those with internal rot, were transferred to pre-labeled straws for drying and processing. Ailanthus leaf tissue was also collected from trees with foliage for genetic analyses as part of a peripheral study . Increment cores were prepared as described by Stokes and Smiley (1996) and Swetnam et al. (1985). Cores were air-dried inside straws for 3–5 days, glued into routed wooden blocks, and taped to secure cores for 2–5 days of additional drying. Intact cores were sanded with an orbital palm sander using 100- and 220-grit sandpaper. Cores were further sanded by hand using a series of sandpapers to 2400-grit, followed by polishing with lamb’s wool to remove fine particulate matter and enhance the wood surface for microscopic examination and ring measurement. Ring-width measurements (0.01 mm) were conducted using a Velmex sliding stage micrometer (Velmex Inc., Bloomfield, NY). Marker years (excessively narrow or wide annual rings) were recorded. Using marker years, individual tree 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 11 Figure 4. Map of study area including detail of Philadelphia, PA, where most trees in the large tree study were sampled. Filled circles without numbers indicate only one tree cored at that given location. Historic garden abbreviations are as follows: BG = Bartram’s Garden, WC = Woodlands Cemetery (formerly The Woodlands), PG = Pratt’s Garden (Lemon Hill), and GR = Grumblethorpe. Highlighted DCNR forest districts (filled gray areas) indicate where Ailanthus were sampled on state and federal forest lands. Counties sampled in state-wide survey not included (see Fig. 13). 12 Northeastern Naturalist Vol. 20, Monograph 10 chronologies were adjusted as needed by inserting locally absent ring values or subtracting false rings (Holmes 1983, Yamaguchi 1991). A minimum core segment length of 50 years was used for cross-dating older trees. For younger Ailanthus trees previously sampled by Schall (2008) on the PA Tuscarora State Forest, a segment length of 30 years was used; these cores were analyzed separately because of their young age. Following initial visual cross-dating, ring-width series for all cores were evaluated using the quality-control program COFECHA (Holmes 1983), a statistical program that calculates series inter-correlations and identifies segments that exhibit cross-dating problems within individual series. Problematic cores that could not be accurately dated were disca rded. Growth responses and patterns in Ailanthus. Ailanthus tree-rings are classified as ring porous with distinct boundaries, a characteristic ideal for tree-ring research (Fritts 1976, Rawlings and Staidl 1924). However, little is known concerning Ailanthus’ radial growth patterns, propensity for ring anomalies such as false and missing rings, or sensitivity to high-frequency variations in annual growth. In addition, dioecious tree species such as Ailanthus have shown differential responses in radial growth to climate and exogenous stress, depending on plant sex (Cedro and Iszkuło 2011, Gao et al. 2010, Iszkuło and Boratyński 2011), emphasizing the need to compare growth patterns between male and female Ailanthus trees. This difference is particularly important since Ailanthus females are prolific seed producers, capable of producing >300,000 seeds/tree annually for sustained periods of time (Bory and Clair- Maczulajtys 1980) and a maximum of a million seeds annually for larger female Ailanthus trees (Illick and Brouse 1926). Such extensive seed production is a sink for carbohydrates and likely impacts stem growth as compared to non-seed-producing (e.g., male) counterparts. We conducted an investigation of Ailanthus throughout PA and surrounding areas, in which we compared annual tree-ring patterns to detect common annual growth signals among Ailanthus trees. All growth series were cross-dated using COFECHA with the previously described methodology. Master chronologies were compared to individual chronologies to assess the strength of common growth signals. Mean sensitivity and mean inter-series correlations were evaluated. Mean sensitivity is a measure of the relative change in ring-width from one year to the next in a given series, whereas inter-series correlation is a measure of the strength of the signal common to all sampled trees at the s ite. Age prediction for Ailanthus To predict ages of large Ailanthus trees with internal rot, which precluded total age measurements, as well as historically documented individuals, we utilized two approaches. First, we explored age-diameter relationships using linear regression analysis, which allowed us to predict total age at DBH, despite internal rot and missing annual rings near the tree center, by using DBH. However, preliminary analyses revealed the method applicable only to open-grown, singlestem trees. Nevertheless, this method proved very useful for predicting year of planting/establishment for open-grown historic Ailanthus trees, using only the 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 13 diameter measurements as reported in 19th- and 20th-century publications, which was subsequently used to help validate historic first reports as well as establish if areas of earliest cultivation were also areas of sustained colo nization. Ailanthus trees were assigned to one of three site-class categories: 1) opengrown: a tree with no competition, 2) forest co-dominant: tree with both inter- and intra-specific competition, and 3) cohort-grown: open-grown group of Ailanthus with intra-specific competition only from individuals within the same cohort. Initial regression analysis revealed no correlation between diameter and age for forest co-dominants, which were then excluded from regression analysis. The remaining Ailanthus trees were further classified as either single or multistemmed. Multi-stemmed individuals included low forking (below DBH) stems, as well as multi-trunk trees that had fused. Multi-stemmed individuals were also excluded from analysis after initial regression output revealed no correlation between diameter and age for this group of trees. In addition, trees >140 cm DBH were excluded from this analysis, since data points from these large trees fell outside the predictable range for which age could be accurately estimated. The second analysis utilized mean ring widths and core measurements to estimate the number of rings in partial cores, where the innermost rings were missing due to rot. A preliminary analysis of 13 complete cores, representing trees from 30–130 cm DBH, revealed that a minimum of 35% of the core was needed to predict age ± 20 years ≈70% of the time (Appendix 3). Preliminary analysis indicated that this method could be used across all site-class categories, as well as single- or multi-stem trees. If usable core length was 75% of the total core length, age ± 10 years could be predicted 77% of the time (Appendix 3). Using a 95% intact core, age ± 6 years could be predicted 100% of the time with exact predictions 30% of the time (Appendix 3). Extant trees were included regardless of site and stem class, unlike the regression analysis. This analysis was then used on extant trees with partial tree cores. Age estimates were calculated by projecting the mean ring width of the partial core across the length of the missing core segment and adding this estimated ring count to the existi ng ring count. Ailanthus floristic surveys Historic surveys and reports: To establish where Ailanthus had been previously known to occur in PA, we conducted an exhaustive query of all known literature and searchable databases to establish a base map from which migration patterns could be hypothesized. County-level surveys. Historically, surveys to determine geographic distribution of flora in PA relied mainly on herbarium specimens (Rhoads and Klein 1993, Wherry et al. 1979). Although this type of survey can accurately determine a plant species’ extant distribution at the time of sapling, it cannot reveal if a species persists over time at a given location. Persistence is especially important in geographic regions where only a few individuals represent the entire population. As a means of circumventing these problems, our study included a county-level survey in which specimens were collected from at least one location within every county where extant Ailanthus occurred. Leaf tissue from male and female 14 Northeastern Naturalist Vol. 20, Monograph 10 Ailanthus and, depending on timing of sampling and availability, mature seed from female trees was collected from 2010–2011 for use in peripheral studies. Forest surveys and site history reconstruction. Although Ailanthus was first introduced into PA in c. 1784, casual observations by foresters suggest this species has only recently become widespread within public forests. To resolve when Ailanthus first invaded these forests, as well as determine the stand history, we conducted a tree-ring study within a number of state and federally managed lands, including one state game preserve, two state parks, seven state forests, the Allegheny National Forest, and US Army Corps of Engineers lands with confirmed Ailanthus populations. At each location, the largest Ailanthus tree, as well as three or four adjacent Ailanthus trees, representative of average-sized stems within the stand, were either felled, from which a cross-section disk was removed at stump level, or sampled at DBH with an increment borer. Our preliminary studies revealed no significant difference between age at stump height and age at DBH. Determination of tree age and growth patterns allowed us to determine probable year when most Ailanthus invaded each site. In addition, it allowed us to determine if Ailanthus had been present in the stand prior to the last major site disturbance, such as timber harvesting, or conversely, if the species invaded the site after the disturbance. Ailanthus stem cross-sections from one location on the PA Tuscarora State Forest that had been previously collected for another study (Schall 2008) were also utilized to add robustness to the dataset. All cross-section and core samples were air-dried and prepared for tree-ring counting as described previou sly. Results Tree-ring data Seventy-five increment cores were extracted from 26 large Ailanthus trees, as well as from 24 additional Ailanthus trees growing across numerous historic locations and sites of interest (Fig. 4). Of these 50 trees, 37 trees were from PA, six from VA, four from DE, and one tree each from OH, NY, and WV (Table 1). Growth and internal decay. Diameters of the 50 Ailanthus trees ranged from 39.1–197.4 cm DBH with an average diameter of 86.1 cm (Table 1). Trees included the overall PA State Champion (HSAa10; Fig. 5A) as defined by the PA Forestry Association and the National Champion (HSAa09; Fig. 5C) as defined by American Forests, National Register of Big Trees. Internal decay (heartwood rot), resulting in partial increment cores, occurred in 16 of 50 (32%) of the old Ailanthus, and averaged 44% decay/tree (% missing wood from extracted core[s]) with a range of 4–89% decay for all Ailanthus containing internal rot (Table 1). Average DBH for trees with detectable internal decay was 106.6 cm compared to 76.5 cm for trees without detectable decay (Table 1). Amount of internal rot within individual Ailanthus trees appeared to be symmetric around the circumference, although a few rotten Ailanthus trees had successive cores that varied by as much as 40% in total length, which justified extraction of multiple cores from some rotten trees, especially when mean 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 15 Table 1. Sampling date, location, size, condition, age, and gene tic relationships for Ailanthus trees sampled in the Northeast US. Year Site Plant Stem DBH Heart Tree IDA sampled General location classB genderC classD (cm)E rot (%) No. rings AgeF HSAa011 2010 Longwood Gardens, Kennett Square, PA 2 A 1 88.1 38 64 ?/101 HSAa021 2010 Woodlands Cemetery, Philadelphia, PA 1 A 2 116.1 73 32 100/? HSAa031 2010 Woodlands Cemetery, Philadelphia, PA 1 A 1 102.6 5 80 88/82 HSAa04 2010 Private residence, Lansdowne, PA 2 A 1 72.1 - 83 83 HSAa051 2011 Private residence, Lansdowne, PA 2 A 1 51.3 - 92 92 HSAa061 2011 Private residence, Lansdowne, PA 2 A 1 61.2 - 97 97 HSAa071 2010 Private residence, Lansdowne, PA 2 A 1 70.1 - 80 80 HSAa082 2010 New York Botanical Gardens, Bronx, NY 1 G 1 85.9 - 79 79 HSAa09 2010 Montross Inn, Montross, VA 1 A 1 197.4 89 37 ?/? HSAa10 2010 Aldie Mansion, Doylestown, PA 1 A 2 144.0 39 74 ?/96 HSAa111,2 2010 Chestnut Hill Community Center, Chestnut Hill, Philadelphia, PA 1 A 1 132.6 - 117 117 HSAa121 2010 Private residence, Society Hill, Philadelphia, PA 1 A 1 129.0 75 22 112/? HSAa132 2010 Dickinson College, Carlisle, PA 1 A 1 100.6 - 80 80 HSAa142 2010 Dickinson College, Carlisle, PA 1 A 1 103.9 - 85 85 HSAa152 2010 Dickinson College, Carlisle, PA 1 A 1 87.9 - 84 84 HSAa162 2010 Dickinson College, Carlisle, PA 1 A 1 93.2 - 79 79 HSAa17 2010 Dickinson College, Carlisle, PA 1 A 1 103.6 62 33 88/85 HSAa181 2010 Sisters of St. Francis of Philadelphia, Red Hill Farm, Aston, PA 2 A 1 71.1 - 57 57 HSAa192 2010 Aldie Mansion, Doylestown, PA 1 A 1 80.0 - 42 42 HSAa20 2010 Raystown Lake, Huntingdon, PA 2 1 39.9 - 48 48 HSAa21 2011 Montross Inn, Montross, VA 1 G 2 87.9 - 60 60 HSAa22 2010 Michaux State Forest, Mont Alto, PA 2 U 1 66.0 - 62 62 HSAa232 2010 Pennsylvania State University, University Park, PA 1 G 1 112.8 13 86 97/97* HSAa242 2010 Fairmont Park, Shofuso Japanese House, Philadelphia, PA 3 A 1 76.5 - 58 58 HSAa252 2010 Fairmont Park, Shofuso Japanese House, Philadelphia, PA 3 A 1 53.3 - 42 42 HSAa261,2 2010 Fairmont Park, Walnut Lane Golf Course 3 A 1 58.4 - 76 76 16 Northeastern Naturalist Vol. 20, Monograph 10 Table 1, continued. Year Site Plant Stem DBH Heart Tree IDA sampled General location classB genderC classD (cm)E rot (%) No. rings AgeF HSAa271,2 2010 Fairmont Park, Walnut Lane Golf Course 3 A 1 62.0 - 77 77 HSAa281 2010 Fairmont Park, W. Sedgewick Street 2 A 1 74.4 - 61 61 HSAa292 2010 Fairmont Park, Lemon Hill Mansion 1 A 1 108.5 - 107 107 HSAa30 2010 Fairmont Park, intersection of Sedgeley and Kelly Drive 2 G 1 65.3 34 78 ?/96 HSAa31 2010 Fairmont Park, Shofuso Japanese House 3 A 1 56.1 25 58 44/77 HSAa321,2 2010 Fairmont Park, Shofuso Japanese House 3 A 1 50.0 - 59 59 HSAa332 2010 Fairmont Park, Shofuso Japanese House 1 A 1 69.9 - 55 55 HSAa34 2010 Bloomsburg Hospital, Bloomsburg, PA 1 G 2 68.8 4 64 56/64 HSAa352 2010 Raystown Lake 3 A 1 39.1 - 47 47 HSAa36 2010 Raystown Lake 2 A 1 43.2 - 41 41 HSAa37 2010 Buchanan State Forest, Franklin County, PA 2 A 1 37.6 - 17 17 HSAa38 2011 Broken Rock Road, Bakerton, WV 2 A 1 98.3 - 79 79 HSAa39 2011 Wilmington Country Club, Wilmington, DE 3 A 1 107.7 79 14 93/? HSAa40 2011 Wilmington Country Club, Wilmington, DE 3 A 1 92.5 21 73 78/90 HSAa41 2011 Wilmington Country Club, Wilmington, DE 3 A 2 89.9 37 63 76/80 HSAa422 2011 Wilmington Country Club, Wilmington, DE 3 A 1 59.4 - 34 34 HSAa43 2011 Private residence, Foster, VA 3 D 2 134.1 - 98 98 HSAa442 2011 Private residence, Foster, VA 3 A 1 85.6 - 96 96 HSAa45 2011 Private residence, Smithfield, VA A 2 127.8 58 67 111/111 HSAa46 2011 Rosemont Manor, Berryville, VA 1 A 2 126.5 - 64 64 HSAa471 2011 Private residence, Lansdowne, PA 2 A 1 131.6 - 105 105 HSAa48 2011 Independence National Historical Park, Indepdence Square, 1 A 1 103.1 49 40 88/75 Philadelphia, PA HSAa49 2010 Fernwood State Forest, Jefferson County, OH 2 U 1 50.8 - 60 60 HSAa50* 2011 State Game Lands #211, Dauphin, PA 2 G 1 39.6 - 41 41 HSAa51* 2011 Bartram’s Garden, Philadelphia, PA - U - - - - - 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 17 Table 1, continued. Year Site Plant Stem DBH Heart Tree IDA sampled General location classB genderC classD (cm)E rot (%) No. rings AgeF HSAa52* 2011 Bartram’s Garden, Philadelphia, PA - U - - - - - HSAa53* 2011 Bartram’s Garden, Philadelphia, PA - U - - - - - HSAa54* 2011 Bartram’s Garden, Philadelphia, PA - U - - - - - HSAa55* 2011 Bartram’s Garden, Philadelphia, PA - U - - - - - HSAa56* 2011 Woodlands Cemetery - U - - - - - HSAa57* 2011 Woodlands Cemetery - U - - - - - HSAa58* 2011 Woodlands Cemetery - U - - - - - HSAa59* 2011 Woodlands Cemetery - U - - - - - HSAa60* 2010 Woodlands Cemetery - U - - - - - HSAa61* 2010 Woodlands Cemetery - U - - - - - HSAa62* 2010 Woodlands Cemetery - U - - - - - HSAa63* 2010 Woodlands Cemetery - U - - - - - HSAa64* 2011 Fairmont Park, Lemon Hill Mansion - U - - - - - HSAa65* 2011 Fairmont Park, Lemon Hill Mansion - U - - - - - HSAa66* 2011 Fairmont Park, Lemon Hill Mansion - U - - - - - ATree IDs followed by a number 1 indicate trees that were cross-dated and used to develop a master chronology for Ailanthus in southeastern PA. Trees followed by a number 2 indicate Ailanthus trees used in regression analysis. Trees HSAa50–61(*) were seedlings or saplings not included in the tree-ring analysis but were part of the floristic studies and were catalog ed in the DNA library for peripheral studies. BSite class refers to environment in which trees were found 1 = open-grown, 2 = forest co-dominant, and 3 = cohort-grown. Dash indicates no data. CAilanthus is androecious (A) or gynoecious (G). Trees of unknown status (U) and dead trees (D) are marked accord ingly. DStem class refers to trees as either 1 = single-stemmed or 2 = multi-stemmed but having a single stem at breast height . EDiameter at breast height (1.4 m above soil line). FAge estimates for trees lacking pith or curvature. The first age estimate is based on linear regression, and the second age estimate is a projection of mean ring width (See methods for details). 18 Northeastern Naturalist Vol. 20, Monograph 10 Figure 5. Examples of large extant Ailanthus included in this study A) HSAa10, Aldie Mansion, Doylestown, PA State Champion Ailanthus, The Pennsylvania Forestry Association; B) HSAa29, Lemon Hill, Philadelphia; C) HSAa09, Montross, VA, National Champion Ailanthus, American Forests, National Register of Big Trees; and D) HSAa23, The Pennsylvania State University, University Park, PA. B and D are the two oldest seed-producing Ailanthus in the US, to our knowledge. Photos taken by M.T. Kasson and L.R. Kasson. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 19 ring width was used for age prediction (data not shown). Regression analysis of percentage internal decay versus DBH revealed a significant positive linear relationship (% internal decay = 0.494[DBH] - 8.8; R2 = 0.391, P = .01). However, DBH accounted for only ≈40% of variability in percentage decay, indicating that other factors contributed to predicting internal decay. Age distribution of intact Ailanthus. Ages for Ailanthus trees, confirmed by ring counts to pith, ranged from 17–118 yrs with a mean of 76 yrs (Table 1). Three trees from PA were >100 years old, including the oldest individual in the study (HSAa11 in Chestnut Hill, Philadelphia), which was 119 years old in 2012 (Figs. 6, 7). To our knowledge, this is the oldest Ailanthus reported within the US or abroad where age determination was supported with tree-ring data. Despite our determinations of advanced age in extant Ailanthus, no individuals exceeded 120 years in age, and therefore could not be directly linked with documented 1stgeneration plantings or establish direct evidence to corroborate ties among early historic planting locations. Nevertheless, the older-aged cohort in and around Philadelphia likely represents naturalized populations established from seed or sprouts of late 18th-century or early 19th-century plantings. However, genetic studies are needed to resolve such relationships. Average age was noticeably less for forest- and cohort-grown Ailanthus (66 and 65 yrs, respectively) compared to open-grown Ailanthus (80 yrs) (Table 1). Seed-producing Ailanthus. Six out of 50 large Ailanthus trees were females, as indicated by presence of seed clusters. Females averaged 76.7 cm DBH as compared to 87.4 for male or sexually ambiguous Ailanthus (Table 1). Two of the sexually reproducing females, including a 107-year-old female tree at Lemon Hill Mansion in Fairmont Park, Philadelphia, and a 97-year-old female tree at The Pennsylvania State University, University Park, represented the oldest seedproducing female Ailanthus trees reported to our knowledge (Fig. 5B, D). As of 2012, seed from the latter Ailanthus tree was viable (M. Kasson, pers. observ.). Growth responses and patterns in Ailanthus General growth characteristics. Annual raw ring width ranged from 0.49– 17.14 mm across all sampled Ailanthus, with a mean annual ring width of 4.23 mm. Rings within the first 10 years of intact cores were excluded from measurements (but not age determinations), since the large rings normally produced in juvenile wood are not normally measured in dendrochronological studies unless first fit to a negative exponential curve (Fritts et al. 1969). Trees from VA and WV had larger average annual ring widths of 5.12 mm, compared to average ring widths of 4.39 mm from PA and NY. The SE PA region chronology possessed relatively strong common signals within the greater Philadelphia area and southeastern PA but not beyond. The SE PA chronology consisted of 15 Ailanthus sampled across 10 geographically distinct locations separated by a maximum distance of 68 km (Table 1, Fig. 8A). Annual raw ring width for these trees ranged from 0.49–17.14 mm with a mean annual ring width of 4.02 mm. Common signals were reflected in the significant inter-series correlation value of 0.350, greater than the critical correlation 20 Northeastern Naturalist Vol. 20, Monograph 10 Figure 6. Photographs of the oldest Ailanthus tree (see arrow) in this study (Aa11, 119 years old in 2012). Photographs show the same tree in 1918 (top) and 2011 (bottom). Tree was 132.6 cm DBH at time of core extraction in 2010. Tree is immediately adjacent to Chestnut Hill Community Centre, 8419 Germantown Avenue, Philadelphia, PA. 1918 photo reprinted with permission of the Chestnut Hill Historical Society. Most recent photo (2011) taken by M.T. Kasson. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 21 coefficient of 0.3281 for the 99% confidence level, using a 50-year segment length. Mean sensitivity was 0.303 (range = 0.145–0.409), a value considered acceptable for tree-ring dating (Grissino-Mayer 2001). Ailanthus trees HSAa29, HSAa30, and HSAa33, all seed-bearing female trees, were excluded from the chronology. Preliminary analysis of increment cores from these individuals revealed that including cores from female Ailanthus significantly weakened inter-series correlations in the SE PA chronology, and their growth was negatively correlated with the master chronology (Fig. 8B). Additionally, a suspected clone of four Ailanthus trees in Fairmont park (HSAa24, HSAa25, HSAa31, and Figure 7. Raw ring width chronologies of the oldest individual Ailanthus trees in this study: A) HSAa11, B) HSAa29, and C) HSAa47. Horizontal line denotes the average ring width for each chronology. Zero values at the far left of each chronology indicate either actual rings that were not measured (filled circles) or estimated rings (hollow circles) based on pith estimators for trees with curved innermost annual rings, but lacking pith. 22 Northeastern Naturalist Vol. 20, Monograph 10 HSAa32), and two other Ailanthus (HSAa19 [Aldie] and HSAa04 [Lansdowne]) were removed due to negative correlation values despite confirmation of the highest correlation predicted by COFECHA. A separate master chronology was developed from 22 Ailanthus trees sampled from a large forested stand, which had established following Lymantria dispar dispar L. (Gypsy Moth) salvage logging in 1983 on the PA Tuscarora State Forest (Schall 2008). This master chronology possessed strong common signals resulting in an inter-series correlation value of 0.540 and mean sensitivity of 0.417 (Fig. 8b). A majority of Ailanthus trees in this stand were ultimately killed by the soil-borne vascular wilt pathogen, Verticillium nonalfalfae Inderb. et al. (formerly V. albo-atrum Reinke and Berthold), which also caused a steady decline in radial growth after 2000 (Fig. 8b; Schall 2008, Schall and Davis 2009a). Cross-dating of Ailanthus chronologies from three geographically separated areas within the primary tree-ring study revealed considerable variability in growth patterns. PA sites included Dickinson College (five Ailanthus trees) and Lansdowne (four trees excluding HSAa05), and a private residence in Foster, VA Figure 8. Raw ring-width master chronologies of Ailanthus trees: A) 15 large Ailanthus throughout southeastern PA, and B) 22 dead and dying Ailanthus naturally infected with Verticillium nonalfalfae in the PA Tuscarora State Forest (Schall and Davis 2009a). Horizontal line denotes the average ring width for each chronology. COFECHA output values are provided in the upper right hand corner for each master chr onology. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 23 (two trees). The four Ailanthus trees at Lansdowne shared strong common signals, resulting in an inter-series correlation value of 0.411 and an average mean sensitivity of 0.311. Five Ailanthus trees at Dickinson College lacked common signals across all segments despite being separated by only <100 m. Although mean sensitivity was 0.368, increment core patterns appeared decoupled from ambient conditions, possibly indicating stand-level variability unrelated to climate. In VA, two trees, identical in age, shared common signals with inter-series correlation value of 0.437 and mean sensitivity of 0.307 yet had diameters that varied by almost 50 cm DBH. This finding again indicates that other external factors were contributing to growth differences. Growth and annual ring anomalies. Although Ailanthus cores are ring porous with obvious tree-ring boundaries, false rings occurred in 30 of 100 examined Ailanthus, including 18 of the 50 large Ailanthus trees and averaged 3.57 false rings/tree (range = 1–15 false rings/tree) for a total of 107 false rings (Table 1). False rings appeared as a dark single row of gum-filled cells (Fig. 9). Although location of false rings within a given growth ring varied, the majority usually occurred near the middle of the annual ring. False rings occasionally occurred adjacent to early wood vessels (Fig. 9D), but were difficult to distinguish in this location (Fig. 9). For seven trees, >2 false rings occurred within the same calendar year on one or more occasions (Fig. 9A, D). Only six trees exhibited false rings in the same year, but common false rings did not occur in trees from the same geographic region, or at the same location within a specific annual ring. Conversely, several rotten trees had multiple partial cores containing false rings, some within the same calendar years and intra-ring locations for two or three cores taken >90° apart, indicating that the false rings likely extended around the entire circumference. More than 90% of the false rings occurred within annual growth rings formed after 1980 (Appendix 4). However, due to significant internal rot near the center of some trees, incidence of false rings could not be determined before 1980 in three of the largest, hollow Ailanthus trees (17%; Appendix 4). Furthermore, a number of false rings may not have been detected because they were not present within the increment core sample. To identify a possible cause of false rings in Ailanthus, we examined trees that had been previously inoculated with a native vascular wilt fungus, Verticillium dahliae Kleb. (Schall and Davis 2009a). This pathogen is known to produce vascular discoloration and induce occlusion in the xylem, without killing infected trees, which could result in false rings. False rings were found present in Ailanthus trees that had been inoculated with V. dahliae, but not in all increment core samples (Fig. 9c). However, stem cross-sections from felled Ailanthus that had been inoculated with V. dahliae exhibited false rings in 100% of the excised cross sections, and only in growth rings that developed after inoculation. We also examined Ailanthus trees that were naturally infected with V. dahliae in a related unpublished study. Some cross-sections from felled trees, as well as some increment cores, contained false rings and yellow discoloration, an internal symptom associated with Verticillium infection (Fig. 9D). The false rings were 24 Northeastern Naturalist Vol. 20, Monograph 10 discontinuous around the tree circumference, which explains the lack of false rings in some increment cores, since the core represents a small percent of the total cross section. Figure 9. Annual growth ring boundaries and false rings (FR) in Ailanthus increment cores. False rings are shown as dark tangential bands (white arrows) within the annual growth ring boundaries and include: A). HSAa12, Chestnut Hill, Philadelphia; B) HSAa03, The Woodlands, Philadelphia; C) Vd-1-06-Aa24, a 2006 Verticillium dahliae inoculated Ailanthus from a previous study (Schall and Davis 2009a); and D) RLK1-11- Aa21, a naturally-occurring V. dahliae-infected Ailanthus. VD denotes vascular discoloration preceding the FR within the same annual growth ring. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 25 Other radial growth anomalies were also revealed from this study. In particular, the chronology from HSAa05, growing near Philadelphia, showed an unusual alternating pattern of narrow/wide annual radial growth from 1927–1965 (Fig. 10). No false rings were apparent in these cores, nor was this pattern detected in adjacent similarly aged Ailanthus trees (Table 1). This tree also exhibited below average growth during 1968–2010 (Fig. 10). The cause of this anomaly is unknown. Differential growth responses in male vs. female Ailanthus. Preliminary crossdating using COFECHA revealed low inter-series correlations values for female Ailanthus as compared to the SE PA master chronology. In addition, inter-series correlation values among seed-producing females were not significant. On the contrary, an attempt to cross-date female trees from the same location, such as HSAa29 and HSAa30, revealed no common signals between the trees, resulting in negative inter-series correlation values. Expanding the chronology to include female Ailanthus outside SE PA further weakened this relationship. Age prediction for Ailanthus Regression analysis revealed a highly significant positive linear relationship (R2 = 0.795, P = 0.0001) between Ailanthus DBH and age (Fig. 11). This relationship held for large, intact trees as well as from reputable historic observations (trees with reported diameters and accompanying photographs and/or descriptions) within PA (Fig. 11). Ailanthus trees included in the regression analysis are Figure 10. Radial growth anomalies in Ailanthus. Tree HSAa05, Lansdowne, PA, showed an unusual alternating pattern of narrow-and-wide annual growth rings from 1927–1965. Single dots on the core indicate decadal marks while double dots indicate the half-century mark. No false rings were apparent nor were these patterns detected in four other large Ailanthus in close proximity and included in this study. In addition, this tree exhibited below average growth from 1968–2010. 26 Northeastern Naturalist Vol. 20, Monograph 10 listed in Table 1. Diameter accounted for nearly 80% of all variability in age for open- and cohort-grown, single-stemmed Ailanthus. Using this regression equation, ages and associated year of origin were estimated for extant, decayed (e.g., hollow) trees (Table 1), as well as previously reported large trees for which only DBH measurements existed (Table 2). Predicted ages for extant Ailanthus ranged from 44–112 years, with the oldest individuals occurring within Philadelphia (112 and 100 years) and southern VA (111 years) (Table 1). These age estimations based on DBH corroborate ages derived from extant, intact Ailanthus. Regression analysis provided age estimates for HSAa02, HSAa12, and HSAa39, which could not be predicted using mean-ring width analysis since the percentage decay exceeded our acceptable limit. For historic trees where diameter measurements were taken directly from published reports, predicted ages ranged from 48–120 years (Table 2). These estimations revealed initial establishment of Ailanthus in six PA counties 70–118 years earlier than previously reported. Furthermore, predicted ages in combination with extant tree-ring data confirmed nearly continuous colonization of Ailanthus at Bartram’s Garden, Philadelphia spanning two centuries and a century or more at several other locations throughout Philadelphia and SE PA including Lemon Hill Mansion, Woodlands Cemetery, and Chestnut Hill (Fig. 12). Although all Ailanthus trees fell within the range of diameters from which accurate age predictions could be made, stem-class could not be determined for many historic trees. This uncertainty may have resulted in over-estimations of age (Table 2), even though multi-stemmed trees represented a relatively small proportion (16%) of the overall extant study population (Table 1). However, Figure 11. Relationship between age and DBH of open-grown, single-stem Ailanthus trees sampled throughout Pennsylvania from 1835–2011. Filled circles denote trees cored for this study during 2010–2011, whereas unfilled circles denote historical observations found in various publications from 1835–1950 (see Appendix 2 for historical observations references). 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 27 Table 2. Estimated ages and year of introduction for historic Ailanthus trees in Pennsylvania using linear regression. Date = date of measurementA. Age = estimated age in years. Year = year of introduction. DBH Tree location Date (cm) Age Year References Bartram’s Garden, Philadelphia 1853 67.8 55 1798 Meehan 1853 1877 90.7 77 1800 Sargent 1878 Humphrey Marshall’s Arboretum, Marshallton (Chester County) 1893 60.0 48 1845 Sargent 1893 Grumblethorpe, Philadelphia 1904 118.1 102 1802 Jellett 1904 Bartram’s Garden, Philadelphia 1919 71.7 59 1860 Anonymous 1919 River Front Park, Harrisburg (Dauphin County) 1924 83.8 70 1854 Phillips 1924 Carlisle (Cumberalnd County) 1925 76.2 63 1862 Illick and Brouse 1926 Five Forks (Franklin County) 1925 103.6 89 1836 Illick and Brouse 1926 Mooredale (Cumberland County) 1925 106.7 92 1833 Illick and Brouse 1926 1925 97.5 83 1842 Illick and Brouse 1926 1925 97.5 83 1842 Illick and Brouse 1926 1925 85.3 72 1853 Illick and Brouse 1926 1925 82.3 69 1856 Illick and Brouse 1926 Pine Grove Furnace (Cumberland County) 1925 91.4 77 1848 Illick and Brouse 1926 Waynesboro (Franklin County) 1925 97.5 83 1842 Illick and Brouse 1926 Private Residence, York Springs (Adams County) 1925 91.4 77 1848 Illick and Brouse 1926 Marshall’s Arboretum, Marshallton (Chester County) 1925 61.0 49 1876 Illick and Brouse 1926 1925 76.2 63 1862 Illick and Brouse 1926 Lutheran Theological Seminary, Gettysburg (Adams County) 1925* 81.8B 68 1857 Lutheran Theological Seminary, Gettysburg, PA Photo Archives 1925* 61.0B 49 1876 Lutheran Theological Seminary, Gettysburg, PA Photo Archives 1925* 61.0B 49 1876 Lutheran Theological Seminary, Gettysburg, PA Photo Archives Bartram’s Garden, Philadelphia 1937 106.7 92 1845 Anonymous 1937 Private Residence., Harrisburg (Dauphin County) 1968 122.2 106 1862 Mickalitis 1969 Private Residence, Steelton (Dauphin County) 1978 127.0 111 1867 Hill 1986 Lebanon Valley Engraving, Lebanon (Lebanon County) 1988** 126.2 110 1878 Clark 1988 Awbury Arboretum, Philadelphia (Philadelphia County) 1988 136.7 120 1868 Clark 1988 Private Residence, Tunkhannock (Wyoming County) 1990 137.4 120 1870 Clark 1991 ADate of measurement sometimes preceed year of publication by several years; therefore, year of publication was used to provide the most conservative estimate of age possible. * denotes estimates from photographs taken between 1911–1925. ** indicates the Ailanthus reported in Lebanon County had a year of most recent measurement in 1900 but appears to be a gross er ror given this observation was published in 1988. BMeasurements were estimated from old photographs and based on c urrent measurements of architectural features (see Fig. 3A). 28 Northeastern Naturalist Vol. 20, Monograph 10 having multiple Ailanthus observations within some counties increases the likelihood that at least one of these trees/county were single-stemmed, resulting in a more accurate estimation of year or decade of establishment (Table 2). In addition, old photographs from some locations also confirmed that many historic Ailanthus trees were single-stemmed (Fig. 3). Mean ring-width analysis, the second analysis used in this study, accurately predicted ages for only extant Ailanthus, with a range of 80–111 years. Ages of two trees were within 5 years of regression estimations, and age of one tree was identical to the regression prediction (Table 1). Mean ring-width analysis provided age estimates for Ailanthus trees HSAa01, HSAa10, and HSAa30, which could not be predicted using linear regression since these individuals did not meet regression site class and/or stem-class criteria (Table 1). In general, mean ring analysis tended to over-estimate age for intact Ailanthus cores compared to regression analysis, but predictions were more accurate for trees <79 cm DBH (Appendix 3). For trees >79 cm DBH, mean ring width predictions were less accurate, except for trees with <25% internal decay (Appendix 3). Furthermore, this over-estimation of age in intact trees increased dramatically as modeled percent internal decay increased. Given the average percentage internal decay for trees with partial cores, emphasis should be placed on regression values rather than ring-width estimations if decay exceeds 25% (Table 1, Appendix 3). Ailanthus Floristic Surveys Historic surveys and reports: Previous literature yielded numerous observations of Ailanthus incidence from its introduction forward to the initiation of Figure 12. Historical accounts of Ailanthus trees found throughout PA from time of introduction through 2012. Vertical white lines represent documented Ailanthus observations. Filled bars (black) denote time periods with confirmed Ailanthus colonization, whereas hashed lines denote time periods without reported observations but with high probability of occurrence. Black triangles denote estimations of establishment and/or planting date based on either linear regression or mean ring-width estimates. Gray bars indicate likelihood of survival beyond publication of the monograph. Other Pennsylvania locations denote both historic and non-historic locations and include observations across ten PA counties (see Appendix 2 for complete list of references) . 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 29 this study in 2010 (Appendix 1, 2). Incidence of Ailanthus increased noticeably between 1820 and 1855, from two extreme SE PA counties to 10 counties forming several geographically distinct populations of Ailanthus throughout the state (Fig. 13). By 1925, many of the scattered geographic pockets of Ailanthus Figure 13. Amended first reports of Ailanthus in PA (1784–2011) by county based on historic accounts, herbarium specimens, recent observations, and dendrochronology. Cross-hatch counties in the 2010 map signify no recent confirmation of Ailanthus since earlier published reports. See Tables 1 and 2 for amendments. 30 Northeastern Naturalist Vol. 20, Monograph 10 had coalesced, adding 17 new counties to the species’ geographic range in PA (Fig. 13). The majority of new counties were in the southern half of the state, where expanded railroad routes formed continuous corridors and likely played a role in Ailanthus seed dissemination. Between 1960 and 2010, reports of Ailanthus in western PA increased noticeably, with the addition of 11 new counties to its range, as well as the addition of six new counties in northeastern PA to its range (Fig. 13). As of 2010, Ailanthus had been reported in 51 of 67 PA counties. Based on previous reports, Ailanthus was largely absent from the northwestern and northeastern regions of PA, likely due, in part, to hard freezes and minimal forest fragmentation or soil disturbance which impeded Ailanthus colonization and expansion (Croxton 1939). County-level field surveys: A total of 135 samples were collected from 60 of the 67 PA counties (Table 1, Appendix 5). Our sampling resulted in first reports of Ailanthus in nine PA counties: Butler, Cameron, Clarion, Clearfield, Columbia, Pike, Somerset, Venango, and Wayne. Ailanthus had not been reported from these counties in previous inventories or surveys (Fig. 14) (McWilliams et al. 2007, Rhoads and Klein 1993, Wherry et al. 1979). Our surveys also revealed that Ailanthus, previously reported in Potter and Wyoming counties within the northern tier of PA, are no longer present and have not been observed since these early reports (Fig. 13). In addition to tissue samples for a DNA library, increment cores were also extracted from Ailanthus trees in Tioga and Bradford counties where Ailanthus had not been reported in previous flora surveys, but was listed after 2004 in FIA inventory data (McWilliams et al. 2007). Tree-ring studies confirmed that Ailanthus had been growing in these two counties for 50 and 35 years, respectively. In both locations, seed-producing females represented the oldest i ndividuals. Trees and saplings along roadways and near urban centers represented >70% of all Ailanthus sampled in PA (Appendix 5), which supports similar findings in WV where Ailanthus distribution was correlated with more urbanized counties (Huebner 2003). Evaluation of several forested areas adjacent to the sampled areas revealed no Ailanthus populations (data not shown), emphasizing the propensity of the species for open disturbed areas. State forest surveys: A total of 38 Ailanthus stands were confirmed across two federally managed forested areas, four state parks, ten state forests, four state game lands, and Fairmont Park in Philadelphia (Fig. 14). Seventeen stands were selected for tree-ring analysis that represented a wide range of Ailanthus ages, densities, management histories, ownerships, and geographic locations (Table 3). Additional Ailanthus stands that contained Ailanthus were identified in the Delaware Water Gap (Eichelberger and Perles 2009), but were not surveyed by the authors. Maximum age of Ailanthus trees in the surveyed stands ranged from 12–91 years across all locations, with a mean age of 36 years (Fig. 14a). Age, incidence, and mean diameter of Ailanthus generally decreased as latitude increased (Fig. 14, Table 3). As expected, trees near Philadelphia were 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 31 Figure 14. Incidence of Ailanthus on PA state and federal lands by forest district, 2010– 2011. Numbers indicate (A) maximum age (living or recently killed) and (B) mean age of living Ailanthus at each location. See Table 2 for stem densities and ages, stand history, and management history and status. Circles in Delaware County, PA (lower right corner) denote Ailanthus trees on private land (HSAa04-07, 47) that were used as proxy due to scarcity of state forest lands near Philadelphia. However, Ailanthus from Fairmont Park (HSAa29) were similar in age to trees at Lansdowne and are reflective of maximum ages in this region (Table 1). 32 Northeastern Naturalist Vol. 20, Monograph 10 Table 3. Location, stand history, and current management stratagies for Ailanthus on PA and federal forests. DBH = average diameter at breast height in cm. Age = max/meanA. Ailanthus Year of State forest district, county/ Most recent Year of No. trees intervention/ General Location Stand history/details disturbance origin present DBH Age strategyB Bald Eagle, Centre Timber harvest between 1966–68. 1968 1960s 2500–5000 8 14/9 2003, H-P BESF, Lickhollow Road Tract Gallitzin, Blair Former farm lands acquired between 1960s–1970s. 1970s 1974 1000–2500 18 36/36 2009, BC-SC Canoe Creek State Park, Water Tower Road Tract Buchannan, Franklin 32-ha clearcut, following 70% mortality from a 1988 1989 5000–10,000 18 22/19 2008, BC-SC BSF, Tobacco Road, Letterkenny Tract multi-year Gypsy Moth infestation. Cornplanter, Warren Natural gas well site (shallow well) immediately Unknown 2009 1––10 <3 cm 1/E 2010, H-E Allegheny National Forest, Warren adjacent to Ailanthus. Gallitzin, Indiana Invasion near edge of parking lot along Rt. 403, Unknown 1957 10––25 <3 cm 53/33 2007, H-E GSF, Clark Run Natural Area, Cramer north of Johnstown. Rothrock, Huntingdon 29-ha clearcut from harvested Pinus virginiana. 1960s 1960 2500–5000 13 51/35 2008, BC-SC Army Corps of Engineers, Raystown Lake, Orchard Road Tract Michaux, Cumberland 40-ha clearcut, Oak Leafroller defoliation. 1966 1974 5000–10,000 18 37/25 2008, BC-SC MSF, Bunker Hill Road Tract Tiadaghton, Lycoming Female Ailanthus planted in 1930s at Camp Kline 2007 1973 50–100 4 33/3 2007, H-P TSF, Pine Creek Rail Trail, Waterville arboretum ajacent to abandoned railroad. Tioga, Tioga 53-year-old female Ailanthus adjacent to river and Unknown 1995 <20 < 1 cm 15/1 2010, H-P TSF, Pine Creek Rail Trail, Blackwell abandoned railroad (tree removed 2010). Weiser, Dauphin 18-ha clearcut, timber sale. Bridge built in 1996 1999 1970 2500–5000 7 41/9 2008, BC-SC State Game Lands #211, Wild Turkey followed by roads in 1997. Timber Tract 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 33 Table 3, continued. Ailanthus Year of State forest district, county/ Most recent Year of No. trees intervention/ General Location Stand history/details disturbance origin present DBH Age strategyB William Penn, Berks Roadside invasion. Unknown 1917, 1987 <50 25 95/24 None, none French Creek State Park, Elverson Forbes, Westmoreland Herbicide killed Ailanthus. 2008 1957 <50 <5 cm 54/3 2008, H-P Donegal Township, Tabernacle Road Tract Rothrock, Huntingdon 3-ha selction cut adjacent to 14-ha clearcut in 1971. 1971 1999 <50 8 12./7 None, None Mule Road Tuscarora, Franklin and Perry 54-ha. salvage harvest of dead oak trees following 1983 1962 10,000 18 49/25 2000, BC-SC TSF, Second Narrows Road Tract Gypsy Moth defoliation. Tuscarora, Mifflin Timber sales by DCNR in 1962–1964 (selection 1997 1971 < 250 13 13/40 2009, BC-SC TSF, Sugar Valley Tract harvest), 1978, 1997–1998 (TSI), and 2001 (salvage shelterwood). Tuscarora, Franklin 1963: first road in this valley built for timber sale. 1986 1970 1000–2500 23 41/28 2006, BC-P TSF, Burns Valley Tract, Franklin Timber sales by DCNR in 1963–1964 (selection County harvest) and 1986 (Gypsy Moth salvage). Tuscarora, Franklin 1970–1973: first road in this valley built for 1989–1990 1971 >10,000 21 37/19 2006, BC-SC TSF, Blue Mountain Tract timber sale. Gypsy Moth defoliation chronic from 1982–1989. Understory dominated by Striped Maple and prospects for desirable regeneration were poor. 1988-1990: salvage clearcut of the area. AMaximum age includes both living and recently killed trees, whe reas mean age is based on age of surviving cohort. BManagement strategies are as follows H = herbicide, BC = biological control (see Schall and Davis 2009a, b), P = persistent (incomplete control by herbicide), SC = sustained control, and E = eradicated. None indicates stands where no control measures have been implemented as of 2011. 34 Northeastern Naturalist Vol. 20, Monograph 10 noticeably older than trees further west and north. At nine separate forest sampling locations, thousands of Ailanthus trees formed continuous singleaged stands indicative of widespread invasion following a large disturbance (Table 3). However, in other stands, older scattered Ailanthus trees had existed years before the last major timber harvest or natural disturbance. Seed from female trees in these stands likely accounted for later invasion of Ailanthus in disturbed sites (Table 3). Salvage logging of oak-dominated hardwood stands, in the aftermath of extensive Gypsy Moth and Archips semiferanus Walker (Oak Leaf-roller)-induced mortality, permitted unprecedented invasion by Ailanthus at four of these locations (Table 3). Other invasions onto state and federal lands resulted from plantings on private residences, adjacent to or on state forest property, where seed was likely disseminated outward from the original planting locations. In addition, PA has leased in-holdings of land on state forests, primarily for hunting camps, for more than a century. Some camps have plantings of exotic tree species for ornamental purposes and/or feeding wildlife. Seed from large seed-bearing Ailanthus trees at some of these camps have led to establishment of Ailanthus within nearby state forests and other public lands (Table 3). These secondary populations now are capable of invading additional state forest lands following future disturbances. Discussion Feasibility of using Ailanthus in tree-ring studies Age and radial growth response. Prior to this study, no thorough evaluation had been conducted to determine the life span of Ailanthus and if Ailanthus trees could be successfully used in cross-dated tree-ring studies. Here we report that Ailanthus reaches ages up to 119 years, which supports a previous report of a 117-year old Ailanthus in England (Cutler et al. 1993) as well as anecdotal evidence of long-lived Ailanthus in PA (Wentz 1950). Greater than 20% of the large Ailanthus cored had, or were estimated to have, 95 or more annual growth rings, especially in urban and suburban environments where life expectancy might actually be enhanced due to limited competition and reduced exposure to pathogens common in forest soils. Therefore, we conclude that Ailanthus can routinely live to >100 years of age. However, concentration of oldest individuals in fragmented urban environments coupled with a high incidence of internal decay may have discouraged previous investigations, resulting in the misconception that Ailanthus trees were short-lived. Illick and Brouse (1926) reported that heartwood rot was most prevalent in Ailanthus trees >90 cm DBH and increased rapidly with diameter, which is consistent with our observations. However, large intact Ailanthus trees were found in high enough frequency to encourage further investigation of this species’ age structure as well as potential for tree-ring analyses. Large trees are still common throughout early points of introduction, including the Philadelphia area, New York City, and Narragansett Bay, RI. The age of large Ailanthus trees in these areas might surpass the maximum life expectancy of Ailanthus determined herein, and may offer additional insight into the maximum life expectancy and growth patterns of this species. A review of Champion 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 35 Tree Registries in other states as well as historic literature, suggests that large old Ailanthus trees may exist elsewhere. Regarding cross-dating, Schall (2008) conducted a preliminary tree-ring study in which he cross-dated a single local cohort of Ailanthus, but did not report COFECHA values. Although there is no standard minimum number of annual growth rings required for successful cross-dating, the reliability of cross-dating often diminishes rapidly for series <40 years in length, and tree-ring chronologies from short-lived species should be treated with caution (Cook and Kairiūkštis 1990). Furthermore, rapidly growing, shade-intolerant species such as Ailanthus often contain numerous growth rings within the juvenile wood. Juvenile wood is the youngest wood near the pith, is strongly influenced by competition, and differs anatomically from mature wood. These factors complicate inclusion of juvenile wood in tree-ring analyses (Worbes 2002). Therefore, previous reports that Ailanthus was short-lived (e.g., 50 years) coupled with its juvenile growth characteristics, may have precluded its consideration for tree- ring analysis. Strong common signals ideal for tree-ring studies were found among old Ailanthus trees within a 68-km radius of Philadelphia, but not beyond. In addition, Ailanthus comprising a single younger cohort of trees within the PA Tuscarora State Forest cross-dated very well despite their young age, suggesting that reliability of cross-dating Ailanthus may occur even at young ages in some stands. Inter-series correlations were statistically significant within both groups, indicating that diameter growth of Ailanthus trees within each of the two locations was responding similarly to exogenous factors such as climate. Successful crossdating of sub-groups of large Ailanthus within this greater Philadelphia area also revealed strong common diameter growth signals within geographically delimited clusters. In contrast, several Ailanthus trees at Dickinson College separated by <100 m lacked common signals, suggesting that either site-specific factors or growth complacency, due to exogenous factors (e.g., fertilization, supplemental watering), likely accounted for the lack of common growth signals. Similar trends were observed in a suspected clone of Ailanthus trees at the Shofuso Japanese House in Fairmont Park where three of four trees growing within 3 m and far removed from other Ailanthus trees lacked common signals, could not be cross-dated, and therefore were not included in the master chro nology. Ripple and Larsen (2000) and Larsen and Ripple (2003) reported that crossdating among Populus tremuloides Michx. (Trembling Aspen) trees was unsuccessful in clonal stands in northern Yellowstone National Park. They reported that vegetative propagation and intra-specific root grafting not only helps maintain discrete aspen populations but may also explain unsuccessful cross-dating among individual trees within a given population. Intra-specific root grafting likely occurs within Ailanthus clones as well (Kowarik 1995). Jelínková et al. (2009) reported substantial interconnectivity within trembling aspen clones via interclonal grafts, which may contribute to growth complacency among individual stems. Similarly, Fraser et al. (2006) observed that less competitive trees gained resources from adjacent trees through root grafts, although the overall impacts of grafts were less evident on aboveground responses such as annual 36 Northeastern Naturalist Vol. 20, Monograph 10 growth rings. Nonetheless, the authors stated that even a moderate transfer of resources to a subordinate tree could reduce competitive asymmetry in grafted trees, especially when resource transfer occurs year after year. Root graft studies with Trembling Aspen, Pinus banksiana Lamb. (Jack Pine), and Pinus contorta Douglas (Lodgepole Pine) reported that age of trees at time of graft initiation varied considerably from 2–90 years, as does time required to complete the graft and duration of the graft (Fraser et al. 2005, Jelínková et al. 2009, Tarroux and DesRochers 2010). These findings involving several species suggest that the influence of intraspecific grafts on individual trees varies in time and space. Despite the potential for growth complacency among Ailanthus at individual sites, Ailanthus trees from the Tuscarora State Forest shared strong common signals. It is likely that these trees seeded into the site following large scale logging, as compared to arising vegetatively from the root systems of harvested parent Ailanthus trees. Such findings indicate that incremental growth sensitivity in Ailanthus may be a proxy indicator for the absence and/or reduced incidence of root grafts. However, studies are needed to test this hypothesis. Growth characteristics and anomalies. Observations of >140 stem cross sections and increment cores collected across the northeastern US confirmed Ailanthus to be ring porous with obvious annual ring boundaries. However, false rings with obvious gum deposits occurred in >20% of trees examined and >40% of large trees cored for age determination, and were especially common within the older trees in the greater Philadelphia area (Appendix 4). Gum deposits and tyloses are common features in vessels and tracheids that have ceased to function, and have been associated with false-ring formation in diseased trees (Christiansen et al. 1999, Ragazzi et al. 2002). In Ailanthus excelsa Roxb. (Indian Tree-of-Heaven), vascular occlusions (gum deposits) develop in response to infection by fungi (Shah and Babu 1986), which supports our observations. Ragazzi et al. (2002) showed that declining Quercus robur L. (English Oak) exhibited gum deposits and false rings from which Fusarium solani f. sp. eumartii (Carpenter) Snyder and Hanson was consistently isolated. Although our results revealed false annual rings within large Ailanthus trees infected with V. dahliae, it is unclear why the incidence of false rings increased dramatically among large non-inoculated Ailanthus after 1980, particularly throughout SE PA (Appendix 4). Unprecedented dieback and mortality of Ailanthus by V. dahliae was reported after 1995 in CT, NY, and VA (Emmerich et al. 1998), which might indicate a change in conditions after 1990 ultimately favored attack of Ailanthus trees by V. dahliae. One plausible explanation for these results might be related to tree age at time of false ring formation. Although false rings were observed during the first 50 years of growth of a few Ailanthus, the majority of false rings occurred in Ailanthus tree-rings formed after >60 years of age. Older trees may be more predisposed to agents such as V. dahliae that may cause such ring anomalies. Otherwise, factors influencing radial growth anomalies were likely due to changing climate phenomena or other abiotic factors beginning in the early 1980s, after which a majority of false rings were formed. Alternatively, trees with ring 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 37 anomalies may represent remaining survivors of a larger Ailanthus population that succumbed to disease from the 1980s–1990s, when Verticillium wilt was observed across the region. Although Verticillium wilt provides a unifying explanation as to the cause of false rings in Ailanthus, it does not explain why the incidence of false ring formation increased after 1980. One plausible explanation for this temporal pattern is the inadvertent introduction of an exotic ambrosia beetle from Asia, Euwallacea validus Eichhoff (formerly Xyleborus validus) (Curculionidae: Scolytinae), which was first detected in the US in NY in 1976 and SE PA (Delaware County) in 1980 (Wood 1977, 1980). Although E. validus has a wide host range in its native China and East Asia, this beetle primarily attacks stressed, dying, or recently dead Ailanthus trees in the US (Schall 2008). Preliminary studies by Schall (2008) and our own studies (M.T. Kasson, M.D. Davis, and D.D. Davis, unpubl. data) revealed that this beetle can passively disseminate Verticillium fungi, which offers an explanation of how previously uninfected stressed trees may have become infected in recent decades. Other growth anomalies (e.g., the alternating pattern of narrow/wide radial growth in HSAa05) were not as widely observed as false rings, suggesting other localized or genetic-mediated phenomena associated with these a nomalies. Age prediction. Regression analysis of open- and cohort-grown, single- stemmed Ailanthus trees revealed a highly significant positive linear relationship between tree diameter and age of intact living trees, as well as ages estimated from diameters taken from reputable historic observations. These analyses support the age/diameter regression curves for aging historic trees as well as rotten extant trees. However, some caveats apply when using this approach. First, extending the diameter-age relationship to large individuals that fall outside the modeled range is not recommended. Exceptionally large diameter trees may represent statistical outliers; therefore, using simple linear regression results in over-estimation of age of these large trees. Paradoxically, these trees often contain the most significant internal rot and/ or may be hollow, precluding accurate age measurements, as was the case for HSAa09, the US National Champion Ailanthus. Secondly, it is unrealistic to assume that diameter-age relationships constructed from data collected among Ailanthus trees in the Northeast would accurately describe the relationship for Ailanthus growing at other locations. Further, age and diameter may not be closely related in some areas, due to competition among individual trees, as well as differing biotic and abiotic stresses, all of which may cause marked differences in growth rates (Cook and Kairiūkštis 1990). Such differences can be observed in even-aged stands having a wide variation of diameters (Cook and Kairiūkštis 1990). For example, within the Buchanan State Forest in south-central PA (location of HSAa37), we observed that a single cohort of 22-year-old Ailanthus trees had diameters ranging from 10–38 cm DBH. Similarly, Espenschied-Reilly and Runkle (2008) reported DBH ranges of >26 cm for 20–25-year-old Ailanthus within a woodlot in Ohio. Such variability in diameter emphasizes the importance of 38 Northeastern Naturalist Vol. 20, Monograph 10 including categorical variables such as stem-class and crown-class within regression models, thus reducing the magnitude of errors. Mean ring-width analysis, the second method used in this study, accurately predicted ages for extant Ailanthus with some overlap in age predictions based on regression analysis (Table 1). Because mean ring-width analysis tended to over-estimate age for intact (non-rotten) Ailanthus cores compared to regression analysis, it is uncertain to what degree pre-existing decay, especially significant levels of decay (e.g., >70% heart rot), affects current ring formation and average growth rates. However, when applied to Ailanthus with <25% decay or in combination with regression analysis in order to make direct comparisons of predicted ages, mean ring width analysis provides a simple and useful analysis for estimating age in Ailanthus trees with internal decay. From cultivation to widespread invasion: A spatiotemporal timeline for Ailanthus Ailanthus has been present in PA for nearly 230 years, a residency time second among invasive tree species only to Acer platinoides L. (Norway Maple), present within the state since c. 1756 (Rhoads and Block 2002). However, Ailanthus has presumably surpassed Norway Maple with regard to invasiveness, fragmenting and disrupting entire native ecosystems. Despite a long residence history, early Ailanthus introductions may not represent the initiation of a sustained, expanding invasion despite continuous colonization in some urban locations such as Bartram’s Garden. On the contrary, sustained widespread invasion by Ailanthus apparently did not occur until the mid- to late 20 th century (Fig. 13, Table 3). To better understand this contradiction, and the implications for management of forest stands that contain the invasive Ailanthus, we must consider the invasion strategy and history of Ailanthus. Ailanthus cultivation and widespread planting did not gain significant momentum until after 1820, when the species became commercially available as an ornamental tree in the eastern US. At the same time, seed-producing (pistillate) Ailanthus trees were first observed, after which the first major expansion of Ailanthus within its naturalized range occurred, likely related to seed production and resultant seedlings (Fig. 14). Similarly, Hamilton’s first-reported introductions (c. 1784) of staminate Lombardy Poplar and Broussonetia papyrifera (L.) Vent. (Paper Mulberry), required a second introduction of pistillate plants to allow seed production and subsequent spread of seedlings (Browne 1846, Downing 1851, Swearingen 2005). Further expansion of the naturalized range of Ailanthus did not intensify in western PA until after 1890, when widespread expansion of railroad systems presumably fostered invasions of weedy invaders such as Ailanthus. For example, the famous railroad Horseshoe Curve near Altoona, PA, was completed by the Pennsylvania Railroad in 1854, effectively traversing the steep Allegheny Mountains and connecting Philadelphia and Pittsburgh. However, widening of the curve to support four tracks did not occur until 1898, after which the volume of train traffic across PA increased dramatically. Concurrently, the incidence of Ailanthus in counties through which main railroad corridors passed also increased. Secondary 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 39 spread (e.g., expansion from founding foci) of Ailanthus in PA was likely facilitated via colonization of railway embankments. The colonization of the open railway corridors, in combination with wind turbulence generated by passing trains, allowed for widespread dissemination of Ailanthus seed and subsequent colonization by seedlings, which is common among other weedy and invasive species (Conolly 1977, Gelbard and Belnap 2003, Harrison et al. 2002). Informal surveys of several railroad corridors by the authors revealed large seed-bearing Ailanthus along railroad grades in Cambria, Dauphin, Lackawanna, Lycoming, and Philadelphia counties, among others, suggesting that railroad corridors continue to facilitate spread. By 1925, the shift from railroads to highways as major means of transportation in the US made roadways important corridors for long-distance dispersal of seed. There was a noticeable increase in both incidence of large (>80 cm DBH) Ailanthus trees, as well as naturalized dense thickets of Ailanthus saplings, along PA highways at this time (Illick and Brouse 1926). Highways continue to play a profound role in spread and growth of invasive plant species, by providing a means of dissemination and an ideal disturbed habitat for seed germination and plant establishment (Mortensen et al. 2009). Significant introduction of Ailanthus onto PA state forest lands did not occur until after 1950. It is likely that some lands purchased by the state during this time period already had established Ailanthus populations, and/or Ailanthus may have invaded state forest lands after purchase. Ailanthus may have existed on some of these lands prior to the 1950s either as single individuals, or as older cohorts with the oldest individuals representing the youngest surviving member of an otherwise extinct cohort. However, the majority of large Ailanthus trees cored during our study were >75 years of age, yet the presence of Ailanthus trees before 1950 were not detected within state forests, based on our increment core data. This finding suggests that the oldest sampled Ailanthus likely represent the first established specimens at each surveyed forest location, which is supported by personal observations of state foresters throughout PA (Table 3). The presence of Ailanthus trees >70 years in age in forests around Philadelphia (Table 1), as well as in Ohio’s Tar Hollow State Forest (Joanne Rebbeck, USDA Forest Service, Delaware, OH, pers. comm.), clearly supports advanced age for Ailanthus within forest settings. Harvesting could explain attrition of older Ailanthus trees in our surveyed forest stands, but the prolific stump and root sprouting that occurs in Ailanthus following cutting would presumably have resulted in higher numbers of similar-aged individuals at the same locations, which was not the case. In addition, previous Verticillium wilt epidemics could have eliminated entire Ailanthus populations within PA state forests. Such wilt epidemics were reported in NYC and, to a lesser extent, Philadelphia in the late 1920s and early 1930s (Beach 1929, Gravatt and Clapper 1932). More recent Verticillium wilt epidemics have occurred in NY in the 1990s (Emmerich et al. 1998) and central PA in the early 2000s (Schall and Davis 2009a). Interestingly, none of the old Philadelphia trees contained false growth rings during the 1930s. False rings may be associated with V. dahliae rather than V. nonalfalfae infections, 40 Northeastern Naturalist Vol. 20, Monograph 10 which suggests several likely possibilities: 1) infected trees ultimately died from the more virulent V. nonalfalfae and were not present in sampled populations; 2) Verticillium infection centers were geographically limited, and therefore not sampled in this study; or 3) false rings are only associated with light Verticillium infections during years when environmental conditions do not favor the pathogen, resulting in survival of infected Ailanthus. The presence of false rings in the oldest Ailanthus at one site in the Tuscarora State Forest offers some credibility to the third hypothesis (Appendix 4). Furthermore, our observations, as well as those of Schall and Davis (2009a, b) have revealed that Ailanthus is much more tolerant to V. dahliae than to V. nonalfalfae. Widespread invasion of Ailanthus on PA state forest lands was generally not initiated until the late-1960s, with most establishment occurring after 1974 (Fig. 13b, Table 3). Although average stand age was not determined for all our sampled stands (Fig. 13), age estimations based on average diameters indicated that most dense Ailanthus thickets were fairly young (Table 3). Stand history records revealed that most (70%) forested stands with >1000 Ailanthus stems had been partially harvested or clearcut immediately prior to Ailanthus invasion. In addition, increment cores extracted from large Ailanthus trees within or immediately adjacent to harvested stands revealed year of establishment generally predated harvest (Table 3). Moreover, some pre-existing Ailanthus trees were bearing seed and were often growing up-wind of recently harvested stands, as with HSAa50 (Table 1, Table 3). Similarly, female Ailanthus trees at Bald Eagle State Forest, Tioga State Forest, and Tiadaghton State Forest produced seed responsible for invasion of nearby logging clear-cuts and abandoned railroad corridors (Table 3). The most significant Ailanthus invasions (including all locations with >5000 Ailanthus stems/location) closely followed large-scale clear-cuts in the aftermath of Oak Leaf Roller defoliation in the late 1960s and early 1970s, as well as subsequent salvage-logging following statewide Gypsy Moth defoliations in the 1980s. Logging operations in these defoliated stands were designed to salvage dead oaks, especially in oak-dominated stands that had suffered repeated defoliation and exhibited considerable oak mortality (Table 3; Herrick and Gansner 1988). Although Ailanthus was likely not present in all stands prior to harvest, the common size and type of harvest apparently led to widespread Ailanthus invasion. These findings suggest that clear-cutting and partial canopy removal associated with salvage-logging operations played an important role in fostering widespread invasion of Ailanthus into the cut-over sites (Table 3). Similarly, Carter and Fredericksen (2007) reported that Ailanthus seedlings tended to dominate recently logged, disturbed sites in southern VA, as compared to undisturbed mature forest stands. Similarly, Martin et al (2010) reported that Ailanthus grew 2.6 times faster than the fastest-growing native species in northwestern CT at 80% full sun but had some of the highest mortality rates in low-light conditions (1% full sun), which would likely limit this species to disturb ed sites. Although Ailanthus seedlings are only occasionally browsed by herbivores such as Odocoileus virginianus Zimmermann (White-tailed Deer), the level of browsing is significantly less in recently logged forests as compared to mature 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 41 forests. This observation is likely related to the high number of Ailanthus stems in clear-cuts, which results in a significant number of non-browsed Ailanthus stems. In addition, heavy logging and clear-cutting result in numerous seedlings of more preferred plant species such as oaks and maples, which are browsed rather than Ailanthus seedlings (Carter and Fredericksen 2007, Williams et al. 2008). The lag between harvest and establishment of Ailanthus within the Bald Eagle State Forest and Michaux State Forest (Table 3) may suggest native regeneration failure followed by subsequent site colonization by Ailanthus seed. This scenario is likely influenced by heavy browsing of non-Ailanthus species in these areas by deer, reducing seedling density and delaying sapling growth of native species, leading to dominance of Ailanthus since this species is not preferred by deer (Augustine and McNaughton 1998, Carter and Fredericksen 2007, Gill and Beardall 2001). Impacts of contemporary and future disturbances on expansion of Ailanthus in PA and beyond The forests of PA are the result of long-term, dynamic stand-history patterns. However, their shared disturbance history has resulted in widespread invasion by Ailanthus and other invasive plant species. Large-scale natural disturbances such as insect defoliation and subsequent tree mortality, wildfire, and canopy gaps foster invasion by non-indigenous plants. In addition, anthropogenic disturbances such as logging, fires, surface mining, railroads and highway construction and corridors, etc. in proximity to established Ailanthus populations exacerbate the naturally occurring disturbances and may lead to widespread invasion across a number of forested sites. These interactions illustrate the importance of implementing preemptive stand-management strategies in our forests prior to overstory removal, as a means of limiting colonization by invasive plant species. Pennsylvania provides a strong case study for other states attempting to develop effective management strategies to limit colonization and range expansion by Ailanthus and other woody invasive species. How Pennsylvania’s forests and similar forests throughout the Northeast respond to contemporary and future disturbances, whether natural or anthropogenic, will depend on the severity and extent of those disturbances. However, the invasion of forests by exotic invasive plant species will not only continue, but will likely increase regardless of management strategy. There are numerous examples of shifts in forest tree species composition due to past insect infestations and disease epidemics. However, contemporary introductions of forest insects such as Adelges tsugae (Annand) (Hemlock Woolly Adelgid [HWA]), Anoplophora glabripennis (Motschulsky) (Asian Long-horned Beetle), and Agrilus planipennis Fairmaire (Emerald Ash Borer), as well as fungal pathogens such as Geosmithia morbida (M. Kolarík, E. Freeland, C. Utley, and Tisserat) and Raffaelea lauricola T.C. Harr., Fraedrich, and Aghayeva, causal agents of Thousand Cankers Disease and Laurel Wilt, respectively, may cause additional widespread changes in the structure and composition of our forests, ultimately opening these stands and favoring invasion by exotic plant species. Orwig and Foster (1998) observed 42 Northeastern Naturalist Vol. 20, Monograph 10 that Ailanthus rapidly invaded Tsuga canadensis (L.) Carr. (Eastern Hemlock) stands exhibiting moderate to severe mortality following HWA infestations in CT. Similarly, hemlock stands killed by HWA in PA, the Mid-Atlantic region, and areas further south will likely experience significant invasions by Ailanthus and other non-indigenous plants. Widespread invasions are likely to occur in areas where large populations of Ailanthus are currently growing in close proximity to devastated hemlock stands such as in the Delaware Water Gap National Recreation Area (DEWA). Indeed, Eichelberger and Perles (2009) reported that numerous hemlock stands at DEWA killed by HWA infestations were susceptible to invasion by exotic plant species such as Ailanthus. These authors also reported that Ailanthus was already present in 20% of DEWA, including areas that contained hemlock communities. Fortunately, many of the intact DEWA hemlock stands are not suitable habitats for Ailanthus, due to limited edge openings, low soil pH, and dense canopies from remaining hardwood associates (Eichelberger and Perles 2009). Yet, declining hemlock stands outside the intact hemlock forests were considered vulnerable to invasion by Ailanthus, especially near roadways and disturbed areas. Furthermore, as hemlock mortality continues, soil pH may gradually increase, favoring future Ailanthus invasions. In support of this contention, we recently observed Ailanthus invading HWA-killed hemlock stands near the Rothrock State Forest adjacent to a major roadw ay. Zomlefer et al. (2008) observed significant mortality of Persea borbonia (L.) Spreng. (Redbay) and Sassafras albidum (Nutt.) Nees (Sassafras) on Cumberland Island, GA, caused by the laurel wilt fungal pathogen, Raffaelea lauricola. Laurel wilt has resulted in significant losses of Redbay and Sassafras, as well as other species, along roadways and within various coastal forest communities. Concurrently, Zomlefer and colleagues (2008) documented Ailanthus invading disturbed areas on Cumberland Island. Although the Ailanthus abundance in the area was relatively low, widespread site disturbance following Redbay mortality will likely provide new opportunities for Ailanthus invasion, especially along roadways where Redbay was a co-dominant tree in the canopy. Although Redbay is not native to the northeastern US, Sassafras is abundant throughout the Mid- Atlantic, where it prefers disturbed sites, abandoned fields, fence rows, and dry ridges and upper slopes (Griggs 1990). Widespread mortality of Sassafras by R. lauricola could favor supplantation by Ailanthus in areas where the two species’ ranges overlap. Exceptional disturbances, such as the catastrophic flooding in PA during 2011 following Tropical Storm Lee, increase the likelihood of riverbanks and flood plains being invaded by Ailanthus. For example, seed-bearing Ailanthus trees identified in our floristic survey were growing in a large cluster along a feeder creek of the Susquehanna River in Bradford County. In September 2011, flood waters downed hundreds of trees including seed-bearing Ailanthus, which were subsequently washed into the Susquehanna River. Large numbers of Ailanthus seeds were transported along the downstream river corridor, which will likely result in establishment of new Ailanthus seedlings along the riverbanks and at other areas within the floodplain. In a related study, Kaproth and McGraw (2008) 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 43 reported that Ailanthus seeds dispersed directly into water can travel long distances and readily germinate, even after prolonged inundation up to 5 months. This study suggests that those seeds swept by the floods will likely survive a prolonged period in floodwaters, allowing Ailanthus seedlings to readily establish on stream banks or in flood plains once waters have rescinded. Some of the most serious impacts currently facing PA forests are from drilling activities to extract and transport natural gas from the Marcellus Shale Formation. Widespread land-clearing, road building, and pipeline establishment related to natural-gas drilling activities have resulted in an unprecedented increase in incidence of disturbed areas ideal for invasion by exotic plants. Considerable Marcellus Shale drilling activities are currently taking place in the northern tier counties of PA, where Ailanthus is currently absent or occurs at low levels. Northern tier counties that contain few if any Ailanthus include McKean, Potter, Susquehanna, and, except for Blackwell, most of Tioga (Fig. 14). Some localized Ailanthus populations have been identified in Bradford and Warren counties, but lack suitable disturbed sites for widespread invasion aside from flood plains and riverbanks. However, natural-gas drilling activities will likely furnish a plethora of disturbed sites ideal for Ailanthus invasion from the few natural seed sources in the area. In addition, some activities associated with the drilling, extraction, and transportation of natural gas require crushed stone or gravel (NYDEC 2009). If gravel is brought in from south of the northern tier counties, there is the added risk that these materials, or the vehicles used to transport them, may harbor seeds from Ailanthus as well as other invasive species, which could become established within the disturbed sites. Movement of new Ailanthus populations into the PA northern tier counties is likely as the connectivity of viable habitat patches increases (Hellmann et al. 2008). This expansion, in addition to the northward migration of tree species such as Ailanthus due to global climate change, may drastically alter the future geographic range of Ailanthus. With regard to climate change, the distribution of Ailanthus in PA clearly shows that few Ailanthus populations have been documented in USDA Plant Hardiness Zone 5a and 5b, a zone comprising the northern tier of PA with average annual minimum temperatures of below -23°C (Fig. 15). The scarcity of Ailanthus in this geographic region of PA, including complete absence of Ailanthus from Elk, Forest, McKean, and Susquehanna counties, which are mainly within Hardiness Zone 5, suggests that the low temperatures, perhaps coupled with minimal forest fragmentation and development are likely limiting Ailanthus establishment and spread in this area (Fig. 15). However, our recent discovery of a 50-year old Ailanthus in Tioga County, as well as the persistence of Ailanthus in Bradford, Crawford, and Warren counties suggests that Hardiness Zone 5b has sustained Ailanthus populations during the past several decades. In contrast, evidence of early 20th-century plantings of Ailanthus in Potter County could not be found, suggesting that conditions unfavorable to Ailanthus (e.g., exceptionally cold winters) probably eliminated this species from these areas sometime after 1925. Our observations of patterns of Ailanthus’ spread further south in PA suggest that incidence of Ailanthus in northern PA will likely increase in close 44 Northeastern Naturalist Vol. 20, Monograph 10 proximity to already established Ailanthus populations at a faster rate compared to establishment of novel outlier populations. Conclusions Ailanthus has a long history in PA, yet widespread invasion by this species was not apparent until the last decades of the 20th century. These observations reflect salvage logging practices following unprecedented mortality of canopy oaks in the aftermath of state-wide Gypsy Moth outbreaks in the early 1980s. Tree-ring studies revealed Ailanthus can exceed 100 years of age, yet many Ailanthus stands, particularly in PA forests, are very young, and maximum age limits are unknown for this species in a natural forest environment. Although no early plantings were found in our recent surveys due to age limitation, extant Figure 15. Historic and contemporary Ailanthus distribution in PA overlaid onto the 2012 USDA hardiness zone map. Ailanthus points include historic observations (gray circles) such as previous flora surveys, herbarium specimens, and from reputable historic publications, as well as our most recent survey (white circles) conducted between 2010–2011. See Appendix 5 for locations of extant Ailanthus. Historic points (gray circles) provided by paflora.org. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 45 individuals were found in a number of historic locations, where tree-rings and associated age-prediction models were able to help resolve spatial and temporal migration patterns and establish areas of continuous colonization within PA including Bartram’s Garden in Philadelphia. Furthermore, these studies support establishment of Ailanthus in several PA counties 70–118 years earlier than previously reported. Although Ailanthus is generally widespread throughout PA, there are still many areas vulnerable to invasion, where cold temperatures and minimally disturbed areas have limited Ailanthus invasion. Although Ailanthus has been eradicated in a few sites, most land managers and private landowners in PA are losing the battle to control the spread of this highly invasive tree species. In the future, land and forest managers should consider various facets of Ailanthus biology and invasion history, as revealed in this monograph, to better formulate management decisions. Acknowledgments The authors thank the PA DCNR Bureau of Forestry and Bureau of State Parks, PA Game Commission, and the US Army Corps of Engineers for their cooperation in locating and granting access to sampling sites, as well as providing study areas for long-term research. In addition, special thanks to Tim Frontz and Rod Whitman who aided in the 2010–2011 Ailanthus floristic survey, Scott Wade, who helped locate large Ailanthus trees and to coordinate introductions to various private landow ners throughout the state, and to many private individuals who allowed us to sample large trees. The authors acknowledge the contributions of Eric Rosko, Eric O’Neal, and Lindsay Kasson in assisting with data collection and increment core extraction. We also thank two anonymous reviewers for valuable suggestions that improved the manuscript. This research was funded by grants from USDA Forest Service Forest Health Technology Enterprise Team (FHTET), Morgantown, WV, and PA DCNR Bureau of Forestry, Harrisburg, PA. Literature Cited Aldrich, P.R., J.S. Briguglio, S.N. Kapadia, M.U. Morker, A. Rawal, P. Kalra, C.D. Huebner, and G.K. Greer. 2010. Genetic structure of the invasive tree Ailanthus altissima in Eastern United States cities. Journal of Botany 2010. Available online at doi:10.1155/2010/795735. Alerich, C.L. 1993. Forest statistics for Pennsylvania: 1978 and 1989. USDA Forest Service Resource Bulletin NE-126, Northeastern Forest Experiment Station, Radnor, PA. 248 pp. Anonymous. 1919. Batram’s Garden property/topographical/planting plan. Early 20th-century linen, 32.5 x 32.25 inches. Held in the Farimont Park Historic Resource Archives. Available online at http://www.philageohistory.org/rdic-images/ view-image.cfm/AS-35. Anonymous. 1936. Planting plan, Batram’s Garden (east side). Coated linen, 28.75 x 19.25 inches. Held in the Farimont Park Historic Resource Archives. Available online at http://www.philageohistory.org/rdic-images/view-image.cfm/AS-37. Anonymous. 1833. The Ailantus [sic]. Independence, Poughkeepsie, NY. 10 July, 1833. P. 1. Augustine, D.J., and S.J. McNaughton. 1998. Ungulate effects on the functional species composition of plant communities: Herbivore selectivity and plant tolerance. Journal of Wildlife Management 62:1165–1183. 46 Northeastern Naturalist Vol. 20, Monograph 10 Barry, P., and J.J.E. Smith. 1855. Arboriculture Notes - No. II: Bartram's Garden, Philadelphia. The Horticulturist and Journal of Rural Art and Rural Taste 5:371–375. Beach. 1929. Diseases of Hardwoods: Ailanthus glandulosa. Plant Disease Reporter Supplements 67–74:372. Belden, L.C. 1958. Humphry Marshall, American quaker botanist. M.A. Thesis. University of Deleware, Newark, DE. 122 pp. Bory, G., and D. Clair-Maczulajtys. 1980. Production, dissemination, and polymorphism of seeds in Ailanthus altissima (Mill.) Swingle, Simarubaceae. Revue Generale de Botanique 88:297–311. [In French]. Browne, D.J. 1846. The trees of America: Native and foreign, pictorially and botanically delineated, and scientifically and popularly described. Harper and Brothers, New York, NY. 520 pp. Carr, R. 1828. Periodical catalogue of fruit and ornamental trees and shrubs, greenhouse plants, &c. cultivated and for sale at the Bartram Botanic Garden, Kingsessing, near Gray's Ferry-Four miles from Philadelphia. Russell and Martien Printers, Philadelphia, PA. Carr, R. 1861. Introduction of the Lombardy Poplar into America. The Gardener’s Monthly and Horticulturist 3:9. Carter, W.K., and T.S. Fredericksen. 2007. Tree seedling and sapling density and deer browsing incidence on recently logged and mature non-industrial private forestlands in Virginia, USA. Forest Ecology and Management 242:671–677. Cedro, A., and G. Iszkuło. 2011. Do females differ from males of European Yew (Taxus baccata L.) in dendrochronological analysis? Tree-Ring Research 67:3–11. Christiansen, E., P. Krokene, A.A. Berryman, V.R. Franceschi, T. Krekling, F. Lieutier, A. Lönneborg, and H. Solheim. 1999. Mechanical injury and fungal infection induce acquired resistance in Norway Spruce. Tree Physiology 19:399–403. Clark, R. 1988. Big trees of Pennsylvania update. Pennsylvania Forests 79:28. Clark, R. 1991. Big trees of Pennsylvania. Pennsylvania Forests 81:13. Conolly, A.P. 1977. The distribution and history in the British Isles of some alien species of Polygonum and Reynoutria. Watsonia 11:291–311. Cook, E., and L.E. Kairiūkštis. 1990. Methods of Dendrochronology: Applications in the Environmental Sciences. Kluwer Academic Publishers, Dordrecht, The Netherlands. 397 pp. Croxton, W.C. 1939. A study of the tolerance of trees to breakage by ice accumulation. Ecology 20:71–73. Cutler, D.F., M.C. Bridge, and P.E. Gasson. 1993. An introduction to dendrochronological work on windblown trees at Kew and Wakehurst Place. Forestry 66:225–232. Darlington, W. 1849. Memorials of John Bartram and Humphrey Marshall with Notices of their Botanical Contemporaries. Lindsay and Blakiston, Phila delphia, PA. 585 pp. Darlington, W. 1853. Flora Cestrica: An Herborizing Companion for the Young Botanists of Chester County, state of Pennsylvania. Lindsay and Blakiston, Philadelphia, PA. 498 pp. De La Hunt, T.J. 1916. Perry County: A History. The W.K. Stewart Co., Indianapolis, IN. 359 pp. Downing, A.J. 1844. A Treatise on the Theory and Practice of Landscape Gardening, Adapted to North America; with a view to the improvement of country residences; comprising historical notices, and general principles of the art: Directions for laying out grounds and arranging plantations, the description and cultivation fo hardy 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 47 trees, decorative accompaniments to the house and grounds, the formation of pieces of artificial water, flower gardens, etc. with remarks on rural architecture. Wiley and Putnam, New York, NY. 497 pp. Downing, A.J. 1851. Trees and pleasure grounds of Pennsylvania. The Horticulturist and Journal of Rural Art and Rural Taste 6:127–130. Downing, A.J. 1852. Shade trees in cities. The Horticulturist and Journal of Rural Art and Rural Taste 7:345–349. Drayton, C. 1806. Personal diary of Dr. Charles Drayton. 2 Novemeber 1806, pp. 49–53. Drayton Hall, National Trust for Historic Preservation, Charleston, SC. Dudley, W.R. 1887. A preliminary list of the vascular plants of the Lackawanna and Wyoming valleys. In Proceedings and Collections of the Lackawanna Institute of History and Science, Vol. 1. The Republica Press, Scranton, PA. 157 pp. Dwight, T. 1845. The Ailanthus. Dwight’s American Magazine and Family Newspaper, New York, NY 1:576. Eastwick, S.W. 1963. The “Grumblethorpe” Garden. The Herbarist 4:11–24. Eichelberger, B.A., and S.J. Perles. 2009. Determining the status and trends of key invasive plant species in the Delaware Water Gap National Recreation Area. National Park Service Technical Report NPS/NER/NRTR - 2009/134, Philadelphia, PA. 61 pp. Emmerich, T., M. Birmingham, and M. Daughtrey. 1998. Naturally-occurring pathogen is killing Ailanthus (New York). Restoration and Management Notes 16:223. Espenschied-Reilly, A.L., and J.R. Runkle. 2008. Distribution and changes in abundance of Ailanthus altissima (Miller) Swingle in a southwest Ohio woodlot. The Ohio Journal of Science 108:16–22. Fraser, E.C., V.J. Lieffers, and S.M. Landhäusser. 2005. Age, stand density, and tree size as factors in root and basal grafting of lodgepole pine. Canadian Journal of Botany 83:983–988. Fraser, E.C., V.J. Lieffers, and S.M. Landhäusser. 2006. Carbohydrate transfer through root grafts to support shaded trees. Tree Physiology 26:1019–1023. Fritts, H.C. 1976. Tree Rings and Climate. Academic Press, New York, NY. 567 pp. Fritts, H.C., J.E. Mosimann, and C.P. Bottorff. 1969. A revised computer program for standardizing tree-ring series. Tree-Ring Bulletin 29:15–20. Gao, L., C. Zhang, X. Zhao, and K.V. Gadow. 2010. Gender-related climate response of radial growth in dioecious Fraxinus mandshurica trees. Tree-Ring Research 66:105–112. Gearhart, E. 1907. The arborescent flora of Lycoming, Northumberland, Snyder and Union counties, Pennsylvania. In report of the Pennsylvania Department of Forestry, for the Years 1905 and 1906, together with Suggestions Concerning the Present Needs and the Future Policy of the Deparment, and Papers Upon Subjects Connected with Pennsylvania Forestry. Harrisburg Publishing Co., State Printer, Harrisburg, PA. 148 pp. Gelbard, J.L., and J. Belnap. 2003. Roads as conduits for exotic plant invasions in a semiarid landscape. Conservation Biology 17:420–432. Gill, R.M.A., and V. Beardall. 2001. The impact of deer on woodlands: The effects of browsing and seed dispersal on vegetation structure and composition. Forestry 74:209–218. Gravatt, G.F., and R.B. Clapper. 1932. Verticillium wilt of maple, Ailanthus, and elm. Plant Disease Reporter 16:96–98. Gray, A. 1868a. Field, Forest, and Garden Botany: A Simple Introduction to the Common Plants of the United States East of the Mississippi, both Wild and Cultivated. Ivison, Phinney, Blakeman, and Co., New York, NY. 386 pp. 48 Northeastern Naturalist Vol. 20, Monograph 10 Gray, A. 1868b. Manual of the botany of the Northern United States. Ivison, Blakeman, Taylor, and Co., New York, NY. 703 pp. Griggs, M.N. 1990. Sassafras albidum (Nutt.) Nees sassafras. Pp. 773–777, In R.M. Burns and B.H. Honkala (Tech. Coord.). Silvics of North America. Vol. 2. Hardwoods. Agricultural Handbook 654. US Department of Agriculture, Forest Service, Washington, DC. Grissino-Mayer, H.D. 2001. Evaluating cross-dating accuracy: A manual and tutorial for the computer program COFECHA. Tree-Ring Research 57:205–221. Hamilton, W. 1809. Advertisement. Poulson's American Daily Advertiser, Philadelphia, PA. 28 August 1809. P. 3. Harrison, S., C. Hohn, and S. Ratay. 2002. Distribution of exotic plants along roads in a peninsular nature reserve. Biological Invasions 4:425–430. Harshberger, J.W. 1899. The Botanists of Philadelphia and their Work. Press of T.C. Davis and Son, Philadelphia, PA. Harshberger, J.W. 1921. The old gardens of Pennsylvania VI. The Woodlands, former Country Seat of William Hamilton. The Garden Magazine 33:120–123. Harvey, M.B. 1895. Something about the Lower Merion. In Historical Sketches: A Collection of Papers Prepared for the Historical Society of Montgomery County, Pennsylvania, Volume 1. Herald Printing and Binding Rooms, Norristown, PA. 418 pp. Hellmann, J.J., J.E. Byers, B.G. Bierwagen, and J.S. Dukes. 2008. Five potential consequences of climate change for invasive species. Conservation Bi ology 22:534–543. Herrick, O.W., and D.A. Gansner. 1988. Changes in forest condition associated with Gypsy Moth on new frontiers of infestation. Northern Journal of Applied Forestry 5:59–61. Hill, J. 1986. Big trees of Pennsylvania. Pennsylvania Forestry Association, Philadelphia, PA. Hobaugh, M. 1993. Big Trees of Pennsylvania. Pennsylvania Forestry Association, Philadelphia, PA. Holmes, R.L. 1983. Computer assisted quality control in tree-ring dating and measurement. Tree-Ring Bulletin 43:69–78. Hu, S.Y. 1979. Ailanthus. Arnoldia 88:29–50. Huebner, C.D. 2003. Vulnerability of oak-dominated forests in West Virginia to invasive exotic plants: Temporal and spatial patterns of nine exotic species using herbarium records and land classification data. Castanea 68:1–14. Huebner, C.D., C. McQuattie, and J. Rebbeck. 2007. Mycorrhizal associations in Ailanthus altissima (Simaroubaceae) from forested and non-forested sites. Journal of the Torrey Botanical Society 134:27–33. Illick, J.S., and E.F. Brouse. 1926. The Ailanthus tree in Pennsylvania. Pennsylvania Department of Forests and Waters, Harrisburg, PA. 29 pp. Iszkuło, G., and A. Boratyński. 2011. Initial period of sexual maturity determines the greater growth rate of male over female in the dioecious tree Juniperus communis subsp. communis. Acta Oecologica 37:99–102. Jelínková, H., F. Tremblay, and A. DesRochers. 2009. Molecular and dendrochronological analysis of natural root grafting in Populus tremuloides (Salicaceae). American Journal of Botany 96:1500–1505. Jellett, E.C. 1904. Germantown old and new: Its rare and notable plants. Germantown Independent-Gazette, Germantown, Philadelphia, PA. John Bartram Association (JBA). 1907. Bartram's Garden, Philadelphia, PA. John Jos. McVey, Philadelphia, PA. 40 pp. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 49 Jones, T.P. 1835. Observations on the Ailanthus tree. Journal of the Franklin Institute October:223. Kaproth, M.A., and J.B. McGraw. 2008. Seed viability and dispersal of wind-dispersed invasive Ailanthus altissima in aqueous environments. Forest Science 54:490–496. Knapp, L.B., and C.D. Canham. 2000. Invasion of an old-growth forest in New York by Ailanthus altissima: Sapling growth and recruitment in canopy gaps. Journal of the Torrey Botanical Society 127:307–315. Kowarik, I. 1995. Clonal growth in Ailanthus altissima on a natural site in West Virginia. Journal of Vegetation Science 6:853–856. Larsen, E.J., and W.J. Ripple. 2003. Aspen age structure in the northern Yellowstone ecosystem: USA. Forest Ecology and Management 179:469–482. Long, T.P. 1991. The Woodlands: “A matchless place”. M.Sc. Thesis. University of Pennsylvania, Philadelphia, PA. 488 pp. Loudon, J.C. 1844. Arboretum et Fruticetum Britannicum; or, The Trees and Shrubs of Britain, native and foreign, hardy and half-hardy, pictorially and botanically delineated, and scientifically and popularly described; with their propagation, culture, management, and uses in the arts, in useful and ornamental plantations, and in landscape-gardening; preceded by a historical and geographical outline of the trees and shrubs of temperate climates throughout the world. Longman, Brown, Green, and Longmans, London, UK. Manning, R. 1830. Ornamental shade trees. Salem Gazette. Salem, MA. 9 April 1830. P. 3. Martin P.H., C.D. Canham, and R.K. Kobe. 2010. Divergence from the growth-survival trade-off and extreme high growth rates drive patterns of exotic tree invasion in closed-canopy forests. Journal of Ecology 98:778–789. McWilliams, W.H., S.P. Cassell, C.L. Alerich, B.J. Butler, M.L. Hoppus, S.B. Horsley, A.J. Lister, T.W. Lister, R.S. Morin, C.H. Perry, J.A. Westfall, E.H. Wharton, and C.W. Woodall. 2007. Pennsylvania’s forest, 2004. United States Department of Agriculture Forest Service Resource Bulletin NRS-20, Northern Research Station, Newtown Square, PA. 86 pp. Meehan, T. 1853. The American Handbook of Ornamental Trees. Lippincott, Grambo, and Co., Philadelphia, PA. 257 pp. Mergen, F. 1959. A toxic principle in the leaves of Ailanthus. Botanical Gazette 121:32–36. Mickalitis, A. 1969. Big trees of Pennsylvania. Pennsylvania Forestry Association, Philadelphia, PA. 31 pp. Miller, J. 1923. Flora of Erie County, Pennsylvania. Printed by the Nature Study Department of the Erie’s Woman Club, Erie, PA. 47 pp. Miller, P. 1835. The Gardener’s Dictionary: Containing the best and newest methods of cultivating and improving the kitchen, fruit, flower garden, and nursery, as also for performing the practical parts of agriculture, including the management of vineyards, with the methods of making and preserving wine, according to the present practice of the most skilful vignerons in the several wine countries in Europe, together with directions for propagating and improving, from real practice and experience, all sorts of timber trees. John and Francis Rivington, London. 722 pp. Minor, C. 1835. Fruit trees, plants, &c. National Banner and Nashville Whig. Nashville, TN. 30 September 1835. P. 1. 50 Northeastern Naturalist Vol. 20, Monograph 10 Monson, A.S. 1843. An address delivered before the New Haven Horticultural Society, May 25, 1843. The Magazine of Horticulture, Botany, and all Useful Discoveries and Improvements in Rural Affairs, Hovey and Co. Boston. 9:346–349. Morse, L.E., J.T. Kartesz, and L.S. Kutner. 1995. Native vascular plants. Pp. 205–209, In E.T. LaRoe, G.S. Farris, C.E. Puckett, P.D. Doran and M.J. Mac (Eds.). Our Living Resources: A Report to the Nation on the Distribution, Abundance, and Health of US Plants, Animals and Ecosystems. US Department of the Interior, National Biological Service, Washington, DC. Mortensen, D.A., E.S.J. Rauschert, A.N. Nord, and B.P. Jones. 2009. Forest roads facilitate the spread of invasive plants. Invasive Plant Science and Management 2:191–199. Moyer, I.S. 1905. Flora of Bucks County, Pennsylvania. Appendix, In W.W.H. Davis’ History of Bucks County, Pennsylvania: From the Discovery of the Delaware to the Present Time. 2nd Edition. Lewis Publishing Co., New York, NY. New York State Department of Environmental Conservation (NYDEC). 2009. Supplemental generic environmental impact statement on oil, gas, and solution mining regulatory program. Available online at http://www.docstoc.com/docs/12365971/ OGdSGEISFull. Accessed 21 December 2011. Oldschool, O. 1809. The Woodlands. The Port Folio 2:505–507. Orwig, D.A., and D.R. Foster. 1998. Forest response to the introduced Hemlock Woolly Adelgid in southern New England, USA. Journal of the Torrey Botanical Society 125:60–73. Overlease, W.R. 1973. The structure of selected deciduous forest in southern Chester County, Pennsylvania. Journal of the Pennsylvania Academy of Science 47:181–197. Palmer, C.R. 1983. Bartram’s Garden, Phila., PA, existing site plan. Hand-drawn map, scale 1" = 30 ft. Held in the Private Collection of the John Bartram Association, Batram Library, Philadelphia, PA. Pavilion. 1830. Ornamental trees. Rhode Island American, Statesman and Providence Gazette. Providence, RI. 30 April 1830. P. 2. Peirce, J. 1833. Nursery at Linnean Hill, near Washington. Daily National Intelligencer. Washington, DC. 9 March 1833. P. 4. Pennsylvania Department of Conservation and Natural Rescources (PA DECNR). 2013. DCNR invasive plants. Available online at http://www.dcnr.state.pa.us/cs/groups/ public/documents/document/dcnr_20026634.pdf. Accessed 26 March 2013. Phillips, H.B. 1924. Some foreign trees used for ornamental planting in Pennsylvania. Forest Leaves 19:104–105. Prince, W. 1827. Extracted from “Prince’s Treatise on Horticulture”. The American Farmer, John D. Toy. Baltimore, MD. 9:308–309. Rabeler, R.K. 1985. Petrorhagia (Caryophyllaceae) of North America. SIDA 11:6–44. Ragazzi, A., S. Moricca, E. Turco, and I. Dellavalle. 2002. Dendroclimatic analysis of Quercus robur infected with Fusarium eumartii. Phytopathologia Mediterranea 41:131–137. Rawlings, F.G., and J.A. Staidl. 1924. The pulping value of Ailanthus. United States Department of Agriculture Volume No. 270, Forest Products Laboratory Madison, WI. Rhoads, A.F., and T.A. Block. 2002. Norway Maple (Acer platanoides L.). Available online at http://www.dcnr.state.pa.us/forestry/invasivetutorial/norway_maple.htm. Accessed 21 December 2011. Rhoads, A.F., and T.A. Block. 2007. The Plants of Pennsylvania: An Illustrated Manual. University of Pennsylvania Press, Philadelphia, PA. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 51 Rhoads, A.F., and W.M. Klein. 1993. The Vascular Flora of Pennsylvania: Annotated Checklist and Atlas. American Philosophical Society, Philadelphia, PA. 636 pp. Riley, C.V. 1874. Sixth annual report on the noxious, beneficial, and other insects of the state of Missouri, made to the State Board of Agriculture, pursuant ot an appropriation for this purpose from the Legislature of the State. Regan and Carter, State Printers, Jefferson City, MO. Ripple, W.J., and E.J. Larsen. 2000. Historic aspen recruitment, elk, and wolves in northern Yellowstone National Park, USA. Biological Conservation 95:361–3 70. Sargent, C.S. 1878. The Ailanthus. Twenty-Fifth Annual Report of the Secretary of the Massachusetts Board of Agriculture:274–282. Sargent, C.S. 1880. Notes on trees in the Arboretum of Humphrey Marshall. The Gardener's Monthly and Horticulturalist 22:16–17. Sargent, C.S. 1888. The Ailanthus. Garden and Forest 1:385–386. Sargent, C.S. 1893. Humphrey Marshall. Garden and Forest 298:461–4 62. Schall, M.J. 2008. Verticillium wilt of Ailanthus altissima. Ph.D. Dissertation. Pennsylvania State University, University Park, PA. 194 pp. Schall, M.J., and D.D. Davis. 2009a. Ailanthus altissima wilt and mortality: Etiology. Plant Disease 93:747–751. Schall, M.J., and D.D. Davis. 2009b. Verticillium wilt of Ailanthus altissima: Susceptibility of associated tree species. Plant Disease 93:1 158–1162. Shah, J.J., and A.M. Babu. 1986. Vascular occlusions in the stem of Ailanthus excelsa Roxb. Annals of Botany 57:603–611. Shouxiang, S., Qixiang, S., and Zemin, W. 2008. The dynamic pattern of Pb concentration in annual rings of 70-year tree of Ailanthus altissima in Hefei city Journal of Chinese Urban Forestry. Available online at http://en.cnki.com.cn/Article_en/ CJFDTOTAL-CSLY200804015.htm. Sinclair, R. 1835. Fruit and ornamental trees, shrubs, &c. Baltimore Gazette and Daily Advertiser. Baltimore, MD. 10 October 1835. P. 3. Smith, B.H. 1905a. Some letters from William Hamilton, of the Woodlands, to his private secretary (continued). The Pennsylvania Magazine of History and Biography 29:143–159. Smith, B.H. 1905b. Some letters from William Hamilton, of the Woodlands, to his private secretary (continued). The Pennsylvania Magazine of History and Biography 29:257–267. Smith, V.A. 2009. Tiny sapling, mighty hope at Independence Hall. The Philadelphia Inquirer 29 October 2009. Smithsonian Institute. 1861. Results of meteorological observations: Made under the direction of the United States Patent Office and the Smithsonian Institution from the year 1854 to 1859, inclusive. Government Printing Office, Washington, DC. 538 pp. Stefferud, A.E. 1949. Trees: The Yearbook of Agriculture, 1949. United State Department of Agriculture, Washington, DC. 944 pp. Stokes, M.A., and T.C. Smiley. 1996. An Introduction to Tree-ring Dating. The University of Arizona Press, Tucson, AZ. Swartley, J.C. 1970. Plan of Bartram’s Garden. Hand-drawn map, scale 1" = 30 ft. Held in the Private Collection of the John Bartram Association, Batram Library, Philadelphia, PA. Swearingen, J.M. 2005. Fact sheet: Paper Mulberry. Plant Conservation Alliance’s Alien Plant Working Group, Washington, DC. 3 pp. 52 Northeastern Naturalist Vol. 20, Monograph 10 Swetnam, T.W., M.A. Thompson, and E.K. Sutherland. 1985. Using dendrochronology to measure radial growth of defoliated trees. Handbook #639. USDA Forest Service, Washington, DC. 39 pp. Tarroux, E., and A. DesRochers. 2010. Frequency of root grafting in naturally and artificially regenerated stands of Pinus banksiana: Influence of site characteristics. Canadian Journal of Forest Research 40:861–871. Thornburn, C. 1830. Ornamental Trees. New-York Spectator. New York, NY. 6 August 1830. P. 3. Turner, G.D., R.J. Van Meter, and G.D. Hertel. 2007. Changes in forest understory composition from 1970 to 2003 at the Gordon Natural Area, an urban preserve in Chester County, Pennsylvania. Journal of the Pennsylvania Academy of Science 81:8–13. Wade, S. 2006. Big trees of Pennsylvania. Pennsylvania Forestry Association, Philadelphia, PA. 32 pp. Wade, S. 2011. Champion trees of Pennsylvania. Available online at http://www.pabigtrees. com. Accessed 26 March 2013. Wardrop, J. 1846. An abridged list of select plants, cultivated for sale by James Wardrop & Co., Manchester nursery, near Pittsburgh. George Parkin, Printers, Pittsburgh, PA. Wentz, A.R. 1950. Minutes of the Annual Meeting of the Board of Directors of the Lutheran Theological Seminary, Gettysburg, PA, May 2, 1950. Lutheran Theological Seminary. 295 pp. Wherry, E.T., J.M. Fogg, and H.A. Wahl. 1979. Atlas of the Flora of Pennsylvania. The Morris Arboretum of the University of Pennsylvania, Philadelphia, PA. 390 pp. Williams, S.C., J.S. Ward, and U. Ramakrishnan. 2008. Endozoochory by White-tailed Deer (Odocoileus virginianus) across a suburban/woodland interface. Forest Ecology and Management 255:940–947. Wilson, W.E. 2011. Mineralogical record biographical archive: Solomon W. Conrad (1779–1831). Available online at http://www.minrec.org/labels.asp?colid=1075. Wood, S.L. 1977. Introduced and exported American Scolytidae (Coleoptera). Great Basin Naturalist 37:67–74. Wood, S.L. 1980. New American bark beetles (Coleoptera: Scolytidae) with two recently introduced species. Great Basin Naturalist 40:353–358. Worbes, M. 2002. One hundred years of tree-ring research in the tropics: A brief history and an outlook to future challenges. Dendrochronologia 20:217–2 31. Wunsch, A.V. 2003. Historic American landscapes survey: Woodlands Cemetery (HALS No. PA-5). National Park Service, US Department of the Interior, Washington, DC. 179 pp. Yamaguchi , D.K. 1991. A simple method for cross-dating increment cores from living trees. Canadian Journal of Forest Research 21:414–416. Zomlefer, W.B., D.E. Giannasi, K.A. Bettinger, S.L. Echols, and L.M. Kruse. 2008. Vascular plant survey of Cumberland Island National Seashore, Camden County, Georgia. Castanea 73:251–282. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 53 Appendix 1. Year of first report of Ailanthus altissima by county and associated references. County YearA ReferenceB Adams 1925 Illick and Brouse 1926 Allegheny 1842 Wardrop 1842 Armstrong 1970 Wherry et al. 1979 Beaver 1915 Archives of the Plant Disease Clinic, The Pennsylvania State University Bedford 1970 Wherry et al. 1979 Berks 1892 The Philadelphia Herbarium at the Academy of Natural Sciences, Philadelphia, PA (uncharacterized, no accession #) Blair 1852 Smithsonian Institute 1861 Bradford 2004 McWilliams et al. 2007 Bucks 1905 Moyer 1905 Cambria 1925 Illick and Brouse 1926 Carbon 1970 Wherry et al. 1979 Centre 1851 Smithsonian Institute 1861 Chester 1820 Darlington, 1853 Clinton 1993 Rhoads and Klein 1993 Crawford 1852 Smithsonian Institute 1861 Cumberland 1925 Illick and Brouse 1926 Dauphin 1924 Philips 1924 Delaware 1852 Smithsonian Institute 1861 Erie 1923 Miller 1923 Franklin 1914 Illick 1914 (Pennsylvania Trees, Bulletin 11) Fulton 1970 Wherry et al. 1979 Greene 2004 McWilliams et al. 2007 Huntingdon 1960 The Philadelphia Herbarium at the Academy of Natural Sciences, Philadelphia, PA Jefferson 2004 McWilliams et al. 2007 Juniata 1970 Wherry et al. 1979 Lackawanna 1887 Dudley 1887 Lancaster 1837 Sargent 1878 Lebanon 1970 Wherry et al. 1979 Lehigh 1938 The Philadelphia Herbarium at the Academy of Natural Sciences, Philadelphia, PA (Accession No. 16659) Luzerne 1905 The Philadelphia Herbarium at the Academy of Natural Sciences, Philadelphia, PA (Accession No. 689227) Lycoming 1905 Gearhart 1907 Mifflin 1889 New York Botanical Garden Steere Herbarium, Bronx, NY Mercer 2004 McWilliams et al. 2007 Montgomery 1893 Harvey 1895 Montour 1970 Wherry et al. 1979 54 Northeastern Naturalist Vol. 20, Monograph 10 County YearA ReferenceB Northampton 1852 US Senate 1864 (Report of the Commissioner of Patents on meteorological observations, 1854–1859) Northumberland 1849 The Philadelphia Herbarium at the Academy of Natural Sciences, Philadelphia, PA Perry 1907 De la Hunt 1916 Philadelphia 1784 Browne 1846 Potter 1925 Illick and Brouse 1926 Schuylkill 2004 McWilliams et al. 2007 Snyder 1905 Gearhart 1907 Tioga 2004 McWilliams et al. 2007 Union 1905 Gearhart 1907 Warren 1891 Illick and Brouse 1926 Westmoreland 1857 Smithsonian Institute 1861 Wyoming 1988 Clark 1991 York 1874 Riley 1874 AYear refers to the the earliest published report for a specific county. Many counties have multiple published reports and herbarium specimens supporting these early dates but were excluded from this table. BReference material includes historic reports and publications, collections from herbaria, recent collections of plant material by corresponding author, and estimates derived using linear regression. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 55 Appendix 2. Location, county, year, and diameter for historic observations on Ailanthus in Pennsylvania. Year = year of observation. DBH Location County YearA (cm)B Reference Aldie Mansion Bucks 1993 144.0 Hobaugh 1993 Awbury Arboreteum Philadelphia 1988 136.7 Clark 1988 Bartram’s Garden Philadelphia 1809 0.0* Browne 1846 Philadelphia 1834 40.4** Breck 1837 Philadelphia 1846 53.3** Browne 1846 Philadelphia 1853 67.8** Meehan 1853 Philadelphia 1855 67.8 Barry and Smith 1855 Philadelphia 1877 90.7** Sargent 1878 Philadelphia 1897 0.0 Wilson 1897 Philadelphia 1907 0.0 JBA 1907 Philadelphia 1919 71.1 Anonymous 1919 Philadelphia 1937 106.7 Anonymous 1937 Philadelphia 1970 0.0 Swartley 1970 Philadelphia 1983 0.0 Palmer 1983 Philadelphia 2010 38.9 M. Kasson, pers. observ. Grumblethorpe Philadelphia 1830 0.0* Jellett 1904 Philadelphia 1904 118.1** Jellett 1904 Philadelphia 1909 0.0 Eastwick 1963 Independence Square Philadelphia 2009 0.0 Smith 2009 Lemon Hill Mansion Philadelphia 1845 0.0 Dwight 1845 Longwood Gardens Chester 2010 88.1 Wade 2011 Marshall’s Garden Chester 1820 0.0* Darlington 1853 Chester 1893 0.0 Sargent 1893 Chester 1899 0.0 Harshberger 1899 Chester 1920 0.0 Harshberger 1921 Chester 1893 0.0 Sargent 1893 Chester 1899 0.0 Harshberger 1899 Chester 1920 0.0 Harshberger 1921 Chester 1925 61.0 Illick and Brouse 1926 Chester 1925 76.2 Illick and Brouse 1926 Chester 1958 87.4 Belden 1958 Woodlands Philadelphia 1785 0.0 Browne 1846 Philadelphia 1790 0.0* Smith 1905 Philadelphia 1921 0.0 Harshberger 1921 Other Philadelphia 1827 0.0* Jones 1835 Philadelphia 1827 0.0* Jones 1835 Philadelphia 1827 0.0* Jones 1835 Philadelphia 1827 0.0* Jones 1835 Adams 1833 0.0 Wentz 1950 Philadelphia 1835 25.7 Jones 1835 Philadelphia 1835 27.4** Jones 1835 Philadelphia 1835 23.9** Jones 1835 Philadelphia 1835 23.4** Jones 1835 Lancaster 1837 0.0* Sargent 1878 Lancaster 1837 0.0* Sargent 1878 Lancaster 1868 45.7** Sargent 1878 Lancaster 1868 45.7** Sargent 1878 Lancaster 1877 51.8** Sargent 1878 56 Northeastern Naturalist Vol. 20, Monograph 10 DBH Location County YearA (cm)B Reference Lancaster 1877 54.9** Sargent 1878 Philadelphia 1904 76.2 Jellett 1904 Philadelphia 1904 0.0 Jellett 1904 Dauphin 1924 83.8 Phillips 1924 Cumberland 1925 76.2 Illick and Brouse 1926 Franklin 1925 103.6 Illick and Brouse 1926 Cumberland 1925 106.7 Illick and Brouse 1926 Cumberland 1925 97.5 Illick and Brouse 1926 Cumberland 1925 97.5 Illick and Brouse 1926 Cumberland 1925 85.3 Illick and Brouse 1926 Cumberland 1925 82.3 Illick and Brouse 1926 Cumberland 1925 91.4 Illick and Brouse 1926 Franklin 1925 97.5 Illick and Brouse 1926 Adams 1925 91.4 Illick and Brouse 1926 Chester 1925 61.0 Illick and Brouse 1926 Chester 1925 61.0 Illick and Brouse 1926 Adams 1925 81.3 Photo Archives of the Lutheran Theological Seminary, Gettysburg, PA Adams 1925 63.5 Photo Archives of the Lutheran Theological Seminary, Gettysburg, PA Other Adams 1925 63.5 Photo Archives of the Lutheran Theological Seminary, Gettysburg, PA Adams 1950 0.0 Wentz 1950 Dauphin 1968 122.2 Mickalitis 1969 Dauphin 1978 127.0 Hill 1986 Lebanon 1988* 126.2 Clark 1988 Philadelphia 1988 136.6 Clark 1988 Wyoming 1990 137.4 Clark 1991 Delaware 2008 71.1 Wade 2011 ADate of observation: the 126.2-cm-diameter Ailanthus reported in Lebanon County had a year of most recent measurement in 1900. This appears to be a gross error given this observation was published in 1988. BAll measurements reported were converted to diameter in centimeters. If measurements were excluded from a specific reference, a value of 0.0 was given. Those 0.0 values followed by * denote a reported planting date at a specific location. **denotes historic measurements used in developing linear regression to predict age based on DBH (Fig. 12). 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 57 Appendix 3. Age prediction validation using two approaches on intact large Ailanthus. Age predictions are by average ring width at several decay intervals (%) A Age by linear Age predictions Tree ID DBH Age regression 85 75 65 50 25 5 HSAa42 59.4 34 48 20 20 29 36 35 32 HSAa36 43.2 41 33 47 52 51 48 43 41 HSAa19 80.0 42 67 40 48 51 50 47 43 HSAa20 39.9 48 29 47 52 49 52 49 51 HSAa32 50.0 59 39 107 96 100 90 71 59 HSAa28 74.4 61 62 47 52 57 58 64 61 HSAa46 126.5 64 110 40 44 51 66 69 64 HSAa27 62.0 77 50 60 84 86 90 81 79 HSAa08 85.9 79 72 60 60 69 88 93 82 HSAa38 98.3 79 84 127 124 117 100 84 76 HSAa13 100.6 80 86 153 140 117 110 100 86 HSAa04 72.1 83 59 87 140 143 128 107 89 HSAa15 87.9 84 74 93 84 83 78 80 84 HSAa44 85.6 96 72 180 172 151 132 113 95 HSAa29 108.5 107 93 80 84 89 108 111 104 HSAa11 132.6 117 116 127 112 117 138 133 119 AMean ring width predictions were determined by counting total rings for each decay class and then taking the average ring width (core width/total rings) to project the number of rings over the length of the missing section. 58 Northeastern Naturalist Vol. 20, Monograph 10 Appendix 4. Incidence and distribution of false rings (FR) in Ailanthus in PA and OHA. Increment No. Year and incidence of false-ring formation in Ailanthus core IDB First ringC FR annual growth increment coresD HSAa02A1 1979 2 1991, 1982 HSAa03A1 1923 15 2008 (2)*, 2007, 2006*, 2004, 2002 (2), 1997, 1996, 1995*, 1993 (2), 1987, 1980 HSAa03B1 1930 7 2006*, 1997*, 1996, 1995 (2), 1994, 1989 HSAa03C1 1973 15 2010, 2008*, 2007, 2005, 2004, 2001, 1998 (2), 1997 (2), 1996, 1993 (2), 1988, 1987 HSAa08A1 1932 1 1974 HSAa09D1 1994 5 2010 (3), 2009 (2) HSAa11A1 1899 8 2009 (2), 1989*, 1988, 1986 (2), 1984, 1980 HSAa11B1 1895 2 2003, 1903 HSAa12B1 1992 2 1995 (2) HSAa18A1 1952 1 1988 HSAa23A1 1923 3 1998, 1997, 1933 HSAa25A1 1954 2 2003*, 1999*, 1998 HSAa27A1 1931 5 1996 (3), 1957 (2) HSAa29A1 1904 2 2010, 2009 HSAa29B1 1910 1 2010 HSAa32A1 1953 1 2004 HSAa35A1 1960 1 1985 HSAa40B1 1939 1 1980 HSAa42A1 1979 1 2008 HSAa43B1 1929 2 1999, 1996 HSAa45D1 1966 3 2009, 2006 (2) HSAa47A1 1907 1 1999 Vd-1-06-232* 1984 2 2009*, 2006* Vd-1-06-242* 1990 2 2006* RLK-443 1966 13 2004, 2002, 2001, 1999, 1997 (2), 1996, 1995 (3), 1987, 1978 (2) RLKb-443** 1976 1 1995 F88-2A4 1960 5 2004, 1995 (3), 1981 F81-1C4 1979 1 2008 GP201-2AB4 1989 1 2008 F-86-1AB4 1973 1 1983 AGeneral location includes Ailanthus trees primarily from large-tree study and three other forested locations where peripheral studies are being conducted. BTrees include individuals from (1) Table 1, (2) Tuscarora SF, Perry County, PA, (3) ACOE Raystown Lake, Huntingdon Co., PA, or (4) Tar Hollow SF, Jefferson Co., OH. Trees were either inoculated (*) with Verticillium dahliae in 2006 or confirmed as having natural infections (**) of Verticillium dahliae in 2009. CFirst ring refers to the year of the first measurable annual gro wth ring for that particular core. DA * following individual ring numbers indicates vascular discoloration preceded false-ring formation. 2013 M.T. Kasson, M.D. Davis, and D.D. Davis 59 Appendix 5. Pennsylvania Ailanthus trees included in 2010–2011 floristic survey. Tree ID TissueA Environment County City/Location Collected Collector P001 Leaf Urban Erie Erie 06/19/10 M.T. Kasson P002 Leaf Wooded Urban Erie Erie 06/19/10 M.T. Kasson P003 Leaf State Park Erie Erie 06/19/10 M.T. Kasson P004 Leaf Urban Venango Oil City 06/19/10 M.T. Kasson P005 Leaf Park Allegheny Pittsburgh 06/20/10 M.T. Kasson P006 Leaf SGL #117 Washington Burgettstown 06/20/10 M.T. Kasson P007 Leaf Forest Huntingdon Raystown Lake 06/21/10 M.T. Kasson P008 Leaf Roadside Bedford Bedford 06/21/10 M.T. Kasson P009 Leaf State Park Forest Blair Canoe Creek SP 06/21/10 M.T. Kasson P010 Leaf Roadside Huntington Mt. Union 06/22/10 M.T. Kasson P011 Leaf Roadside Perry East Salem area 06/22/10 M.T. Kasson P012 Leaf Forest Dauphin SGL #211 06/22/10 M.T. Kasson P013 Leaf Roadside Franklin Mont Alto 06/23/10 M.T. Kasson P014 Leaf Forest Franklin Buchanan SF 06/23/10 M.T. Kasson P015 Leaf Forest Somerset Boswell 06/23/10 D.D. Davis P016 Leaf Roadside Pike Matamoris 06/26/10 M.T. Kasson P017 Leaf Urban Lackawanna Scranton 06/26/10 M.T. Kasson P018 Leaf Urban Cambria Johnstown 06/26/10 D.D. Davis P019 Leaf Roadside Indiana Blairsville 06/26/10 D.D. Davis P020 Leaf RR tracks Westmoreland Seward 06/26/10 D.D. Davis P021 Leaf Urban Mifflin Lewistown 06/28/10 M.T. Kasson P022 Leaf Roadside Juniata Port Royal 06/29/10 M.T. Kasson P023 Leaf Roadside Clearfield Clearfield 07/01/10 D.D. Davis P024 Leaf Roadside Jefferson Punxsutawney 07/03/10 D.D. Davis P025 Leaf Roadside Armstrong Sagamore 07/14/10 D.D. Davis P026 Leaf Forest Lycoming 07/21/10 E.S. O'Neal P027 Leaf Roadside Luzerne Dupont 07/24/10 M.T. Kasson P028 Leaf River bank Warren Warren 07/21/10 R.P. Long P029 Both PA turnpike Lebanon 07/30/10 M.T. Kasson P030 Both Roadside Adams Gettysburg 07/30/10 M.T. Kasson P031 Both Urban York York 07/30/10 M.T. Kasson P032 Leaf PA turnpike Berks 07/30/10 M.T. Kasson P033 Both Roadside Delaware Chester 07/30/10 M.T. Kasson P034 Both PA turnpike Montgomery 07/30/10 M.T. Kasson P037 Seed Roadside Chester 07/30/10 M.T. Kasson P038 Leaf Roadside Cumberland 07/30/10 M.T. Kasson P039 Both Roadside Cumberland Camp Hill 07/30/10 M.T. Kasson P040 Leaf RR tracks Lancaster Downingtown 07/30/10 M.T. Kasson P041 Seed Industrial Lancaster Downingtown 07/30/10 M.T. Kasson P042 Leaf Forest Chester West Chester 07/30/10 G. Hertel P043 Leaf I-80 E Clinton 08/12/10 M.T. Kasson P044 Both I-80 E Snyder 08/12/10 M.T. Kasson P045 Both I-80 E Clinton 08/11/10 Tim Frontz P047 Both Urban Northumberland Mt. Carmel 08/12/10 M.T. Kasson P048 Leaf Roadside Lehigh Allentown area 08/12/10 M.T. Kasson P049 Leaf Roadside Schuylkill 08/12/10 M.T. Kasson P050 Leaf Roadside Monroe Snyderville 08/12/10 M.T. Kasson P051 Leaf Roadside Northampton Nazareth 08/12/10 M.T. Kasson P052 Leaf I-80 E Montour 08/12/10 M.T. Kasson P053 Both Roadside Union Lewisburg 08/12/10 M.T. Kasson 60 Northeastern Naturalist Vol. 20, Monograph 10 Tree ID TissueA Environment County City/Location Collected Collector P054 Leaf Urban Carbon 08/12/10 M.T. Kasson P055 Leaf Roadside Cameron 08/06/10 T. Frontz P056 Leaf Roadside Butler 08/09/10 T. Frontz P057 Leaf Forest Beaver 08/20/10 E. Rosko P058 Leaf I-80 E Montour 08/31/10 T. Frontz P059 Leaf Substation Crawford Meadville 09/02/10 T. Frontz P060 Both Roadside Lawrence 08/30/10 T. Frontz P061 Leaf Roadside Union 08/31/10 T. Frontz P062 Leaf Roadside Clarion 09/01/10 T. Frontz P063 Leaf I-80 E Mercer 09/02/10 T. Frontz P064 Both Roadside Northumberland Milton area 08/31/10 T. Frontz P065 Seed Roadside Centre Bellefonte 09/20/10 M.T. Kasson P066 Both PA turnpike Beaver 09/23/10 T. Frontz P067 Leaf Roadside Fayette Normalville 10/06/10 D.D. Davis P068 Leaf Forest Delaware Aston 10/12/10 M.T. Kasson P071 Leaf Roadside Perry Millerstown 10/15/10 M.T. Kasson P072 Leaf Roadside Perry Newport 10/15/10 M.T. Kasson P073 Leaf Roadside Fulton McConnellsburg 10/15/10 M.T. Kasson P074 Leaf Forest Franklin Buchanan SF 10/30/10 M.T. Kasson P076 Leaf Frick Park, Forest Allegheny Pittsburgh 11/07/10 M.T. Kasson P077 Leaf Forest Huntington ACOE forest 11/03/10 M.T. Kasson P078 Leaf Forest Huntington ACOE forest 11/03/10 M.T. Kasson P082 Leaf Rail trail Lycoming Waterville 11/15/11 M.T. Kasson P083 Leaf Rail trail Tioga Blackwell 11/15/11 M.T. Kasson P084 Seed Riverbank Bradford Athens 11/17/11 M.T. Kasson P085 Leaf Roadside Wayne 06/26/10 M.T. Kasson P086 Seed Roadside Greene 08/08/11 S. Coons P087 Seed Roadside Greene 08/08/11 S. Coons S020 Seed Forest Franklin Tuscarora SF 09/04/07 M.J. Schall S038 Seed Forest Franklin Buchanan SF 10/09/09 M.T. Kasson W01 Wood Roadside Indiana Cramer 10/9/11 M.T. Kasson W02 Wood forest Centre Bald Eagle SF 11/10/11 M.T. Kasson W03 Wood forest Berks French Creek SP 11/17/11 M.T. Kasson W04 Wood Forest Huntingdon Rothrock SF 11/10/11 M.T. Kasson W05 Wood Forest Cumberland Michaux SF 11/10/11 M.T. Kasson W06 Wood Forest Mifflin Tuscarora SF 11/10/11 M.T. Kasson W07 Wood Forest Perry Tuscarora SF 12/2/11 M.T. Kasson W08 Wood Forest Mifflin Tuscarora SF 12/2/11 M.T. Kasson W09 Wood Forest Franklin Tuscarora SF 11/10/11 M.T. Kasson ATissue refers to specific tissues collected at each location. Both refers to leaf plus seed. DNA was extracted from both leaf and seed tissue for a related study. Cross-sections and cores were sampled for tree-ring analysis (see Fig. 14).