The New York Botanical Garden was established in 1891 by an act of the State Legislature of New York to pursue a mission of botanical research, education, and horticulture. In 1895, founding director Nathaniel Lord Britton selected a 100-hectare site in the Bronx, New York’s northernmost borough, as the Garden’s location. Britton chose the site because it included a 20-hectare (50-acre) old-growth forest surrounded by fields with soils suitable for growing a diversity of plants to serve the Garden’s research and educational activities. Even as they were assembling living collections of plants from around the world, the Garden’s founders were documenting the spontaneous flora of the landscape. Early inventories of the Garden’s plant collections listed wild plants as well as cultivated specimens. Botanists deposited vouchers of wild plants collected in the Garden and adjacent Bronx Park in the Garden’s Herbarium. Native trees were identified and labeled for the public’s benefit.

Elizabeth Gertrude Knight Britton, noted bryologist and wife of Nathaniel Britton, took a particular interest in the native plants of the Garden. In the early 1900s she noted that populations of certain native species in the Garden were declining precipitously due to a variety of factors. Her concern about the fate of native plants at the Garden and beyond inspired her to become a founding member of the Wildflower Preservation Society.

The spontaneous flora of The New York Botanical Garden has changed significantly since the Britton’s time. As the Bronx went from a rural or suburban environment to a completely urban one, and as areas of the Garden were devoted to ornamental plantings, the proportion of native and non-native species changed in largely predictable patterns that have been observed throughout the region and indeed throughout the world. Introduced pathogens decimated once robust populations of American chestnut (Castanea dentata) and American elm (Ulmus americana) in the early 1900s, eastern hemlock (Tsuga canadensis) in the 1980s, and currently threaten our native ashes (Fraxinus species).

Determining the spontaneous flora of the grounds of a botanical garden presents a number of unique challenges. It is not always clear if a given plant is cultivated or spontaneous, particularly if it occurs in areas that were once actively cultivated but may be fallow. Some cultivated plants, particularly annuals, reoccur from year to year but only in or around locations where they are regularly planted. Some plants persist and spread vegetatively but do not appear to form spontaneous new populations. Some plants appear to be spontaneous but are present in such low numbers that their long-term viability as naturalized populations is uncertain.

In spite of these challenges, documenting the spontaneous flora of the Garden’s dynamic landscape serves many purposes. It provides invaluable information for horticulturists at the Garden and throughout the region, who can use these data to identify potential new invasive species. This information will improve regional efforts to prevent future invasions. These data also document biodiversity for ecological studies, regional floras and conservation efforts.

In 1916, Nathaniel Lord Britton and Elizabeth Gertrude Knight Britton argued that The New York Botanical Garden and other green spaces should serve as refuges for native plants in the City (E. K. Britton, 1916; N. L. Britton, 1916). In spite of significant stresses inherent in its urban location and substantial anthropogenic change to its spontaneous flora, the Garden remains a refugium for native species and perhaps because of these stresses, is a compelling location for botanical and ecological research.

The percentage of native (62%) versus non-native spontaneous species and infraspecies (38%) extant in the Garden today compares favorably with other natural areas in the region, including New York's Central Park with sixty percent non-native (DeCandido et al., 2007), Pelham Bay Park with sixty-one percent native (DeCandido, 2004) and Concord, Massachusetts with sixty-one percent native (Primack et al., 2009). Of course there are many factors complicating a comparison such as the total area covered by the flora, site history, location, and study methods. However, the data presented here are a first approximation and guide for research and conservation efforts.

Site description

The New York Botanical Garden (40.862N, 73.877W) is set on 100 hectares (250 acres) in Bronx County, New York. Designated a National Historic Landmark in 1967, the Garden features 50 curated gardens, displays, and plant collections and extensive natural areas including the 20-ha (50-acre) old-growth Thain Family Forest, the Bronx River and its floodplain, Twin Lakes, and the Mitsubishi Wild Wetland (Rachlin et al., 2007). Approximately 70 hectares (175 acres) of the landscape is intensively cultivated, paved, or otherwise developed. In addition to the spontaneous flora documented here, the Garden’s living collections include more than one million plants representing more than 14,000 taxa growing out of doors or under glass.

The Garden is located in the southwestern corner of the New England Physiographic Province, near the confluence of New England, Atlantic Coastal Plain, and Piedmont Provinces, and is characterized by warm humid summers and cold winters with periodic drought and regular severe storms (N. L. Britton, 1913; Anonymous, 1938, 1944). The mean average annual temperature measured at Central Park 12 km (10 mi) southwest of the Garden is 12.72° C (54.90°F) with a January average of 0.33° C (32.59°F) and a July average of 24.72° C (76.50°F) (NWS, 2014). The annual precipitation is 126.85 cm (49.94in), evenly distributed throughout the year (NWS, 2014).

The geology of the Garden includes all of the primary bedrock formations of New York City: Fordham Gneiss, Inwood Marble and Walloomsac Schist west of the Bronx River; Manhattan Schist in the center of the Garden; and the Hartland Formation in the eastern portion of the landscape (Merguerian, 2013). Bedrock throughout the Garden exhibits evidence of glaciation, including glacial striations, glacial erratics, and glacial potholes. The highest point is 43 m (141 ft) at the top of a rock outcrop in the center of the Forest and the lowest point is 10 m (32 ft) where the Bronx River passes beneath the Linnaean Bridge at Fordham Road (LaFave et al., 2003). The soils of upland areas consist largely of Charlton, Chatfield, and Hollis acidic sandy loams with varying depths to bedrock (Shaw et al., 2007). In wetlands and along the Bronx River floodplain, soils consist of Canandaigua (fine-silty), Tonawanda (coarse-silty), and Natchaug (loamy) silt loams or organic soils (Shaw, 2010). Recent studies have revealed that the Garden soils have a pH range of 3.9–8.0; have high concentrations of heavy metals including zinc, lead, and copper; are compacted due to excessive trampling; and exhibit hydrophobic properties due to long-term deposition of hydrocarbons from the burning of fossil fuels (White & McDonnell, 1988; Schuler, 2006, 2011; Gabel, 2014;).

Nathaniel Lord Britton and the Garden’s founding Board of Managers chose the Garden’s location in 1895 because the site included many features considered ideal for the establishment of an institution devoted to the study and display of plants: accessibility, rich and varied soil, beautiful scenery, and the largest remnant of natural forest in New York City (Britton et al., 1895). From the beginning, Britton and his scientific colleagues recognized that the wild plants growing in the Garden’s landscape could serve the institution’s mission. The 1896 “General Plan of The New York Botanical Garden” developed by Britton and landscape architect John Brinley with input from Calvert Vaux and Samuel Parsons, preserved the Forest, known then as the “Hemlock Grove,” in its entirety (Britton, 1904). Britton also decreed that the many mature native trees that shaded the old fields around the Hemlock Grove should be protected during the construction of roads and buildings. The Forest today is a 20 hectare (50 acre) uneven aged, mixed hardwood, remnant urban old-growth forest that has never been significantly cleared and effectively remains a natural ecosystem (McDonnell, 1988; Shaw et al., 2007; Loeb, 2011).

There have been many changes to the spontaneous flora of the Garden since the late 1800s. As early as 1902, Garden scientists documented that the over-collection of native wildflowers (e.g., Arisaema triphyllum and Epigaea repens) was leading to their decline (Britton, 1902; Copp, 1904; Britton, 1912a, 1912b; E. K. Britton, 1913). Chestnut blight (Cryphonectria parasitica [Murrill] Barr) led to the demise of 1500 Garden individuals of American chestnut in the early 20th century (Anonymous, 1911). Eastern hemlock, which represented 36% of the canopy trees in the Forest in 1937, has been nearly eliminated by two invasive insects: elongate hemlock scale (Fiorinia externa Ferris) and hemlock wooly adelgid (Adelges tsugae Annand) (Rudnicky & McDonnell, 1989; Schuler & Forrest, 2016). The trampling of soil by crowds of visitors early in the Garden's history was observed and mitigated by fencing the paths in the Hemlock Grove (Britton, 1904).

Although extensive research indicates that the Forest has never been cleared and remains an effectively natural ecosystem in the middle of one of the world’s most developed urban areas, its spontaneous flora has changed significantly since 1895 (McDonnell, 1988; Shaw et al., 2007; Loeb, 2011). In 1937 the most abundant canopy species based on density were Tsuga canadensis, Quercus spp., and Betula lenta (Rudnicky & McDonnell, 1989). In 2011 the most abundant canopy species based on density were Quercus spp., Prunus serotina, and Acer rubrum (Schuler & Forrest, 2016).

As the Garden’s native plant populations have changed, the number of naturalized species in the flora has increased. Some, such as Ficaria verna and Reynoutria spp., have become invasive and pose a serious threat to remaining populations of native plants. Recent efforts to control the spread of invasive species and re-establish populations of extirpated species in the Thain Family Forest indicate that some of the changes to the Garden’s spontaneous flora may be reversible (Schuler & Forrest, 2016).

The boundaries of the Garden have shifted since 1895 (Fig. 1). In 1915, land east of the Bronx River including the former Lorillard Mansion and 56 hectares (140 acres) south of the Stone Mill were acquired (Britton et al., 1915). In 1937, approximately 48 hectares (120 acres) of land in the northeastern part of the Garden, including the historic shrub and willow collections and a portion of the Bronx River and its floodplain, were relinquished for the expansion of the Bronx River Parkway (Robbins, 1939; Robbins, 1941).

Fig. 1
figure 1

Map of New York Botanical Garden. Prepared by Wayne Cahilly, Lionel Goldfrank III Manager, NYBG Institutional Mapping Department.

Full size image

Materials and methods

Names of flowering plants, ferns, and mosses growing wild at the time the Garden was incorporated were published in a series of appendices to the Annual Reports in the Bulletin of The New York Botanical Garden in 1898, 1899, and 1900. The publications are unsigned and there are no introductions or explanation as to how the surveys were conducted, nor, more importantly, whether specimens were collected and preserved. Only in the case of the mosses is there any explicit mention of collections (Nash, 1900).

The first list (Britton, 1898) is an inventory of all plants "in the Grounds" of the Garden, including cultivated plants in nurseries, greenhouses, herbaceous grounds, borders, bog gardens, etc. The wild plants are marked with a "W", nursery plants with "N", greenhouse plants with "G" and so on. The list consists of binomials and trinomials alphabetical by genus and species with no families, authors, nor specimens cited. In the second list (Britton, 1899), those marked as "W" from the 1898 list were extracted and published under the heading "Wild Flora" apart from sections detailing the plants under cultivation. The list is nearly the same as the first list of binomials and trinomials but with the names grouped by family and arranged systematically by kingdom, subkingdom and family. Again, the names do not have authors and no specimens are cited. Some species were added to the list (e.g., Acer saccharinum, Cuscuta gronovii, Geranium maculatum, Ilex verticillata, Linum virginianum, Vaccinium corymbosum). In 1900, George V. Nash, the head gardener, published a list of additions to the 1899 list. This third list comprises mosses and 22 ferns and flowering plants (Nash, 1900).

After publication of the early Garden flora, only occasional accounts of various groups of wild plants (ferns, shrubs, and trees) were produced, some published and some not. From 1917 to 1920, Nash produced a series of works on the "Hardy Woody Plants in The New York Botanical Garden (Nash, 1917–1920). John Kunkel Small and Edward Johnston Alexander later published separate lists of native ferns (Small & Alexander, 1933a; Small, 1934), shrubs (Small & Alexander, 1933b), and trees (Small & Alexander, 1933c). The list of shrubs includes fifty species of native shrubs they report as "self-maintained" as well as thirty-five "Old World shrubs naturalized in the Garden". For the most part, however, their lists do not indicate which are naturalized and which are planted and they do not cite specimens.

Between 1950 and 1986, collecting on the grounds of the Garden was very sporadic. Not until after Michael Nee arrived at the Garden in 1986, was there an effort to systematically document the spontaneous flora. Nee collected numerous specimens on the grounds always indicating whether a species was likely spontaneous or cultivated. Nee kept a working manuscript of the Garden flora, adding species and observations as they were encountered.

From 1986 to 2015 and throughout all seasons, we explored the Garden grounds identifying and collecting those species not yet documented or exhibiting unusual morphological or phenological traits (such as unusual flowering times, invasiveness, etc.). We combined all names from the historic lists (Britton, 1898, 1899; Nash, 1900) and updated the nomenclature to modern equivalents. We then searched the herbarium for all these names as well as all names reported for Bronx County by the New York Flora Association (Weldy et al., 2015) and Taylor (1915). Specimens labeled "Bronx Park", "New York Botanical Garden", or variations of "Hemlock Grove" in the Bronx were identified, databased, imaged and cited here. Specimens marked as "cultivated" are not cited here. All specimens cited are housed at NY, unless otherwise noted. Species that persist from cultivation and spread vegetatively, but do not form new populations, such as Hosta lancifolia Engl., Hyacinthoides hispanica (Mill.) Rothm., Narcissus pseudonarcissus L., Pachysandra terminalis Siebold & Zucc., and Toona sinensis (A. Juss.) M. Roem. are not treated here. Rare escapes from cultivation found in and immediately around greenhouses or propogation areas are not reported here.

The historic and extant spontaneous flora (Appendix I) is arranged by major groups (ferns, gymnosperms, basal angiosperms [magnoliids], monocots and dicots) with families, genera and species arranged alphabetically. Fern taxonomy follows Smith et al. (2006). Flowering plant family taxonomy follows the classification of the Angiosperm Phylogeny Group (APG III, 2009). Author abbreviations follow Brummitt and Powell (1992). Misapplied names and synonyms from Britton (1898, 1899) and Nash (1900) are cited in braces with the species where the material applies.

New taxa named from spontaneous Garden plants (Appendix II) were found by searching the Garden's type database as well as literature searches. New Records (Appendix III) were found by searching the New York Flora Atlas for taxa not reported for Bronx County and comparing the list to the historic and extant flora (Appendix I). Rare plants (Appendix IV) were found by searching the New York State Department of Environmental Conservation's Active Inventory List (NYS DEC, 2016). Only extant species are reported as rare. Waifs as defined here are plants of anthropogenic origin that appear to be spontaneous but occur only as solitary individuals or only very localized populations. Some species collected on the grounds of the Garden were added to the New York State flora (Weldy et al., 2015), but were collected only once and apparently never spread further (e.g., Cynosurus echinatus and Rorippa indica). These are treated here as waifs. Some may be persistent from cultivation and are spreading vegetatively or they may be spontaneous but there is no definitive evidence indicating their status (e.g., Convallaria majalis and Hemerocallis fulva). Our approach is conservative. Some species treated here as waifs are local escapes from plants cultivated on the grounds (e.g., Cephalotaxus harringtonia and Ilex opaca). Others are escaped from plants cultivated elsewhere in the region (e.g., Cleome hassleriana Chod.). Sight records (Appendix V) are those species for which we found no voucher specimen. These taxa are not used in counts or analyses.

Vernacular names are from Gleason and Cronquist (1991), The New York Flora Atlas (Weldy et al., 2015) and USDA NRCS Plants Database (USDA NRCS, 2016). They are provided for convenience. As is the case with the majority of Cyperaceae and Poaceae, these names are contrived and may not be used by any living person familiar with the plants. Frequency values are site-specific for the Garden generally for the period between 1986 and 2015, but principally for the latter years. Frequency values are not meant as regional assessments. The categories used here are as follows: Historic–once present, but not found between 1986 and 2015; Rare–scarce, less than 2 colonies; Infrequent–uncommon, occasional, 2 to 5 colonies; Frequent–common, more than 5 colonies. A species may be reported as "Historic" even though the species is cultivated today. The species Campsis radicans (L.) Seem. ex Bureau. was reported as wild (without voucher) in the early flora (Britton, 1899), but today is known only from cultivation.

All analyses (counts of taxa, new records, lost and gained taxa, etc.) are based on current classification.

Results

The largest extant families are Asteraceae (40 species or infraspecies), Cyperaceae (35), Poaceae (32), Rosaceae (19) and Polygonacaee (18). The largest extant genus is Carex with 27 species or infraspecies, followed by Persicaria with ten, Viburnum and Cyperus each with seven, and Quercus with six (Table I).

Table I The historic and extant spontaneous flora as reported in Appendix I. Numbers are total species and infraspecies. Numbers in parentheses are percentages, rounded to the nearest whole number. A. Total spontaneous flora as reported by Britton (1899) and Nash (1900). B. Number of taxa from A not found among the extant spontaneous flora (F). C. Number of species and infraspecies persistent from A to F. D. Number of taxa from F not present in A. E. Taxa not present in A and not present in F. F. Species and infraspecies extant in 2015. G. Total number of species and infraspecies from 1899 to 2015 (A+D = G).
Full size table

Among the rare extant species and infraspecies (Appendix IV) on the New York State Department of Environmental Conservation Active Inventory List (NYS DEC, 2016), three are critically imperiled (S1) and two are imperiled (S2). The extant species Carex aggregata is currently listed as historic (not seen in 20–30 years) in New York State (NYS DEC, 2016), though we discovered it at the Garden in 2009. Some species, not rare in New York State, are nonetheless uncommon in New York City such as Maianthemum canadense (Canada Mayflower). This species still forms dense patches in the Thain Family Forest. Of the extant native species and infraspecies, 124 are currently rare on the grounds (47%). The families with the greatest number of locally rare extant taxa are Cyperaceae (15 rare species and infaraspecies), Poaceae (12) and Ericaceae (8). Many groups have experienced significant losses (e.g., native Orchidaceae, native Scrophulariaceae, and native Fabaceae). All tree species remain extant with the exceptions of Castanea dentata and Quercus coccinea. The Yellow Birch, Betula alleghaniensis Britton was not reported on any historic list, but apparently appeared spontaneously.

Seventeen families were gained since 1900 (Athyriaceae, Cystopteridaceae, Salviniaceae, Woodsiaceae, Ceratophyllaceae, Berberidaceae, Bignoniaceae, Boraginaceae, Caprifoliaceae, Elaeagnaceae, Grossulariaceae, Lythraceae, Malvaceae, Mazaceae, Menispermaceae, Nyctaginaceae, Paulowniaceae). Among those species and infraspecies gained since 1900 (283), native and non-native, the largest families are Asteraceae (38), Poaceae (22), Rosaceae (21), Brassicaceae (16) and Cyperaceae (14).

Among the 412 species and infraspecies reported in 1900, the largest families are Poaceae (64) species and infraspecies, Asteraceae (41), Cyperaceae (36), Rosaceae (13), and Fabaceae (11). The largest genus in 1900 was Carex with 28 species and infraspecies, followed by Dichanthelium with 13 and Ranunculus and Solidago with eight each. Eleven families reported in 1900 are no longer extant (Acoraceae, Aquifoliaceae, Aristolochiaceae, Cistaceae, Melanthiaceae, Myricaceae, Ophioglossaceae, Phrymaceae, Polygalaceae, Santalaceae, and Saxifragaceae). The largest groups of native species and infraspecies lost since 1900 (181) are Poaceae (29), Asteracae (19), Cyperaceae (15), Ranunculaceae (10), and Rosaceae (7). Some common weeds present in 1900 are no longer extant (e.g., Achillea millefolium, Elymus repens, Phleum pratense, and Saponaria officinalis). In 1900 there were 12 pteridophyte species reported for the Garden (Britton, 1899; Nash, 1900). Today there are 12 extant pteridophytes. Five were lost (Dryopteris cristata, Dryopteris intermedia, Equisetum fluviatile, Equisetum hyemale, and Sceptridium dissectum) and five were gained (Athyrium angustum, Azolla caroliniana, Cystopteris tenuis, Dryopteris carthusiana, and Woodsia obtusa). Some species were spontaneous in 1900, but are now only known from cultivation (e.g., Geranium maculatum, Juglans nigra, and Lobelia cardinalis) and are reported as Historic. In some cases we found no historical material collected from the grounds, but often found specimens from nearby areas. In the case of Osmunda regalis which is reported on the 1898 and 1899 lists, we found material from McLeans Woods (Holtzoff s.n.), Riverdale (Bicknell 11426a) and Pelham Ave (Burnham 701), but nothing from "The New York Botanical Garden" or "Bronx Park".

The nomen nudum "Lobelia Canadensis" was reported by Nash (1900). In the Nash document, the name is listed with Lobelia cardinalis. These two were added to two other Lobelia from the grounds of the Garden (Lobelia inflata and Lobelia siphilitica), previously reported by Britton (1899). We searched the NY herbarium thoroughly and unsuccessfully for any specimen bearing the name "Lobelia Canadensis."

Discussion

All floras are inherently dynamic, subject to biotic and abiotic influences. In natural systems, floristic change usually occurs slowly (over thousands or millions of years) or episodically as a result of cataclysmic events like floods, landslides, droughts, etc. In contrast, recent studies show that modern, urban floras have undergone rapid and dramatic floristic change since the industrial revolution, particularly during the 20th century (Robinson et al., 1994; Pysek et al., 1995; Bertin, 2002; Standley, 2003; DeCandido, 2004; McKinney, 2006, 2008; DeCandido et al., 2007; Primack et al., 2009). Urban floras face unprecedented pressures including climate change, the heat-island effect, acid rain, air and water pollution, deposition of heavy metals, habitat destruction, construction, competition from introduced species, vandalism, soil compaction, etc.

Studies show a marked increase in non-native species and a decline in natives for urban ecosystems of our region. Staten Island as a whole lost almost 41% of its native species from the period 1879–1991 (Robinson et al., 1994). Concord, Massachusetts, an area with a long tradition of conservation has lost 27% of its native species since 1837 (Primack et al., 2009). Central Park has lost 70% of its native species since 1857 (DeCandido et al., 2007). Small populations like those that occur on islands or in isolated patches in urban areas have been shown to be more susceptible to extirpation than larger ones (Robinson et al., 1994). Williams et al. (2009) identified three sources of species in urban environments (native species extant in the environment; native species occurring regionally; non-native species introduced naturally or through human agency). Restoration efforts using native plants add a fourth category of species that may introduce or re-introduce viable species to an urban flora.

Factors contributing to changes in species composition within the Garden include the above regional factors as well as the following factors. Loss of territory of course meant a smaller area for plants. The Garden lost the northern floodplain forest in 1938, eliminating a large area suitable for silver maple (Acer saccharinum) and other species. Change in land use from farmland, pasture, homesites and industrial operations eliminates disturbance regimes particular to these activities, such as mowing for hay versus mowing for lawn. Some were probably brought in from more rural areas with livestock and machinery. The alteration of habitat such as grading and construction eliminated areas where plants could grow. Predation from rabbits, rats and other animals impact the vegetation. Gathering of plants by the public; trampling and soil compaction and ground fires impacted the vegetation in the past.

In 1900, much of the Bronx was rural with farms and scattered homesites. There was a Bedford Park bog with Dichanthelium, Pogonia ophioglossioides, Rhexia, and other bog plants. In her book, Bog Trotting for Orhids, Grace Greylock Niles (1904) described the area with abundant wildflowers: "Along the higher ridges, the brilliant Rock Pinks (Phlox subulata L.) bloom abundantly. Their mossy-mats creep over the hills from Bronx Park to Yonkers." The abundance of Poaceae and Asteraceae in the historic flora suggests that open meadows were a prominent feature of the landscape. Indeed, hay was being cut and sold for revenue within the grounds until at least 1897 (Britton, 1899). Land west of the Bronx River and adjacent to the Metro North railroad tracks (formerly the Harlem division of the New York Central and Hudson River Railroad) at the time of incorporation was described as "mainly open fields which had been under cultivation up to the time that the Bronx Park was established…" (Eustis, 1902: 105). Into the 1920s there was a meadow behind the Museum building with typical meadow species such as Achillea millefolium, Coreopsis lanceolata, Silphium perfoliatum, and Tragopogon pretensis. Other meadow species collected on the grounds early in the Garden's history include Asclepias incarnata and Gentiana andrewsii.

Some species were not recorded from the grounds during the 1898–1900 surveys, but were almost certainly present (e.g., non-natives: Morus alba, Verbascum thapsus; natives: Populus deltoides, Persicaria punctata, Persicaria pensylvanica, Persicaria lapathifolia, Rubus allegheniensis, and Podophyllum peltatum). On the other hand Allium tricoccum Aiton was collected from the "vicinity" (e.g., McLeans Woods, Williams Bridge, Bedford Park and Riverdale) and was cultivated, but apparently never wild collected on the grounds of the Garden. Similarly, it is surprising that Lysimachia borealis (Raf.) U. Manns & A. Anderb. (Primulaceae) has never been found on the grounds when it has been collected on Long Island, Staten Island, and Westchester County. The orchid flora of The New York Botanical Garden originally consisted of just five vouchered species: Calopogon tuberosus, Corallorhiza maculata, Platanthera hookeri, Spiranthes cernua, and Spiranthes lacera. This is surprising because the greater New York City metro area historically contained 45 species. A couple of possible hypotheses may explain the relative paucity of orchids. First, most native orchids are edge or open growing plants that do not compete well in wooded or shrubby habitats. However, regionally common woodland species such as Cypripedium acaule Aiton or Goodyera pubescens (Willd.) R. Br. have never been collected from what eventually became the Thain Family Forest. Second, as noted above, much of the land surrounding and comprising the grounds of the Garden were meadows, where many orchids regionally common to such habitats, e.g., Platanthera lacera (Michx.) G. Don or Platanthera psycodes (L.) Lindl., were collected. Thus, this paucity of orchids may be due not to a deficiency of suitable habitat, but may instead be due to overcollection by the local population prior to the establishment of the Garden, although it is impossible to verify this hypothesis. Today, the only spontaneous orchid to occur within the Garden is Epipactis helleborine, a common and regionally expanding Eurasian orchid typically found in disturbed garden beds, sidewalk cracks, and wooded edges. Why this non-native species continues to maintain healthy populations, vs. the extirpation of all five native species, is an intriguing question with wider implications for the native flora of urban refuges.

Some species were collected before 1898 but not reported on the early lists: Actaea pachypoda was collected in "Woods, Bronx Park" in May 1891 (S. Clarke s.n.) and again by Charles Gilly (444) in 1940 from "New York Botanical Garden, Bronx Park"); Agrimonia rostellata was collected by W. Clute & P. Wilson (s.n.) on 1 September 1899; Amsonia tabernaemontana was collected in a "field near greenhouses" in 1893 (S. Clarke s.n.), but not reported on the early list, and collected again in 1940 from "Bronx Park" by Charles Gilly (311) and again "in waste grassy clearing, not cultivated" by Harold Moldenke (20593) in 1950; Cardamine parviflora was collected from "Fordham Hemlock woods" by Eugene Bicknell (4454) in 1881; Rosa palustris was collected by George Nash (s.n.) in Bronx Park in 1896; Lespedeza repens was collected by E. Bicknell (5170 and s.n.) in September 1891; Vicia sativa was collected in "Bronx Park" May 1891 by S. Clarke (s.n.).

Most curiously, perhaps, is the addition of Picea mariana to the wild flora of the Garden by Nash in the supplement to the larger list (Nash, 1900). This is a boreal species at the southern end of its range, occurring in bogs and frigid swamps. We found several specimens gathered from cultivated trees dating from the period, but none from wild or spontaneous individuals. The species is not formally excluded here from the Garden flora. If accepted generally, the native occurrence of this species in the Bronx is a county record (albeit based on historic occurrence).

The changed garden boundaries complicate comparison between the historic flora and the extant flora, especially when attempting to compare species per unit area for the historic and modern periods. Because species were not collected continuously throughout the Garden's history and the early collectors often only reported the locality as "Bronx Park" it is impossible to know what species occurred in exactly what area at what time. However, for the modern period (1986 to 2015), we did attempt to document every species with precise localities and day of collection. So for the modern period and for the current boundaries, it is possible to calculate species per unit area.

The Garden was established during a dynamic period in botanical science, especially in the area of nomenclature. The "American Code of Botanical Nomenclature" was being debated and Nathaniel Britton was a leading proponent. Eventually, the American Code became the basis for the code used today, but the process was dynamic and contentious. The early Garden flora was published during this period of flux in plant nomenclature and employed novel nomenclatural concepts such as tautonyms (e.g., Benzon benzoin, Malus malus, Opuntia opuntia, Sassafras sassafras) that were dropped from the adopted code. Britton and colleagues were meticulous classifiers of taxonomic groups, often recognizing segregate genera long lumped under the Linnean system. Later workers such as Gleason and Cronquist (1991) again lumped some of these species into larger genera (e.g., Corydalis sempervirens for Capnoides sempervirens, and Prenanthes trifoliolata for Nabalus trifoliolatus), which are once again being recognized in a narrower sense, largely on the basis of molecular evidence in addition to the morphological data available in 1900.

Historic names not in use today were traced through the successive editions of the "Britton and Brown" floras and correlated with contemporaneous specimens from the area whenever possible. In most cases the synonomy was straightforward and unambiguous (e.g., Ailanthus glandulosa became Ailanthus altissima). In some cases, however the disposition of a name has not been straightforward. Contemporaneous specimens from "Bronx Park" (G. Nash 319) and "Bronx Meadow" (E. Bicknell 10312) labeled as Panicularia fluitans and Glyceria fluitans, respectively, are in fact specimens of Glyceria septentrionalis. We put "Panicularia fluitans" in braces with Panicularia septentrionalis, although the name is actually a synonym of Glyceria fluitans (Soreng, 2003), which is also reported here with the synonym Panicularia brachyphylla Nash in braces. Another example is Malva rotundifolia which was reported as wild at by Britton in 1898. In the 1913 edition of the Britton and Brown flora the species is reported as "common nearly throughout our territory" and depicted with smooth mericarps (Britton & Brown, 1913). In the 1952 edition (Gleason, 1952), the name as we know it today, Malva neglecta, is said to be "included in Malva rotundifolia of Gray, B & B., Small, Rydb." The name is actually misapplied to both Malva neglecta and Malva pusilla, but more commonly to the former. The latter is more common westward and is not currently known from New York (Weldy et al., 2015). The species Malva rotundifolia does occur in the region, but is uncommon and has rugose mericarps. We found no historic specimens of wild material from the grounds of the Garden annotated with the name Malva rotundifolia, however, there are specimens of Malva neglecta collected by Britton and others from the New York region annotated as Malva rotundifolia (e.g., Bicknell 5769 from Van Cortlandt Park). We applied the name to Malva neglecta.

Some species were listed twice under two different names in the Britton (1899) list. For example, both Poa flava and Tricuspis seslerioides are listed under Poaceae. Both names are synonyms of Tridens flavus and are listed as such in the Britton and Brown flora of 1913. Botrychium dissectum was reported by Britton (1899) and later Nash (1900) reported Botrychium obliquum. Today these are considered synonyms of Sceptridium dissectum and are so treated here. Carex lurida and Carex tentaculata are both listed in the 1899 list. The latter is a synonym of the former and is listed as such in the Britton and Brown flora of 1913. The same is true for Carex costellata and Carex virescens, both of which are reported on the 1899 list and were later listed under Carex virescens in the 1913 Britton and Brown flora. Sisyrinchium angustifolium and Sisyrinchium graminoides E. P. Bicknell are both recognized by Britton (1899). Today we view the species as a single, if somewhat variable entity (Sisyrinchium angustifolium). However, these examples are few and the vast majority of names on the historic list correspond to valid species today (though sometimes with different names).

Some of the names from the historic list are based on misidentifications or alternate application of a name. Hylodesmum nudiflorum (L.) H. Ohashi & R. R. Mill., reported as Meibomia nudiflora (Britton, 1898), is based on a misidentified specimen. The Bicknell specimen (s.n.) from "Bronx Park", collected on 20 September 1896, is Desmodium perplexum. The name Eragrostis purshii was applied to a specimen collected on 1 September 1899 by W. Clute and P. Wilson (s.n.). The specimen is Eragrostis pectinacea, a species persisting in one locality on the grounds today. Lespedeza frutescens reported by Britton (1898) is based on misidentified specimens of Lespedeza violacea.

Nineteenth Century specimens labeled "Bronx Park" could include plants actually collected from what is now the Wildlife Conservation Society (Bronx Zoo) or areas of Bronx Park now north of the Garden (see Fig. 1), although most were collected after 1896 when the Garden was established. No taxa were added to the spontaneous flora (Appendix I) solely on the basis of a specimen labeled "Bronx Park". All historic taxa are either reported by Britton (1898, 1899), Nash (1900) or explicitly cited as "New York Botanical Garden" on preserved specimens.

We cite all voucher specimens seen for a particular taxon from the flora area, rather than just one "representative specimen" as is sometimes done. If only one specimen per taxon is cited and that specimen is later determined to be misidentified, it is not possible to exclude the name because there is always the possibility other, correctly identified specimens could exist. If we have similarly erred in the application of a name, future workers will be able to confidently exclude that taxon from the flora because we have cited all specimens seen by us that are the basis for that report. The same holds true for application of synonyms and misidentifications.

Species reported by Britton (1899) and Nash (1900) that are not known to form spontaneous new populations but may spread vegetatively in a very localized area are treated here as waifs: Hemerocallis fulva, Nepeta cataria, and Ornithogalum umbellatum. These names are not counted in the historic flora. The report by Britton (1898) of Syringa vulgaris L. as wild is certainly an error. The species is only known to be persistent from cultivation. We are not aware of any credible reports of spontaneous lilacs anywhere in the region and indeed, none was collected. The species is excluded here. Populus alba, also long-persistent from cultivation, but never spontaneous was listed by Britton (1899). No spontaneous voucher was found and it is excluded here. Hibiscus syriacus is also a long-lived shrub or small tree and may appear spontaneous, but we have seen no truly spontaneous populations. Like the preceeding, no spontaneous vouchers were found, and the species is excluded here.

The relative abundance of a species often fluctuates from year to year, depending on many factors including climactic variation, disturbance, etc. For example, in some years, in some areas, Persicaria extremiorientalis may be abundant in a location and scarce or absent in others (Atha et al., 2010). In a very few cases (e.g., Achillea millefolium or Poa pratensis), a species is reported from the historic list and has undoubtedly been here ever since, but we have seen no modern voucher specimen from the grounds of the Garden. These species are cited here as historic, though they are likely extant today. Rorippa indica persisted as a weed on the grounds of the Garden at least from 1947 to 1949, but has since vanished from New York State (Weldy et al., 2015). The species is treated here as a waif.

Plants from the Garden’s living collection have sometimes escaped cultivation and become established in the spontaneous flora of the landscape. Viburnum plicatum forma tomentosum was documented as naturalizing in the Forest in 1935 (McLean, 1935). Today, Ficaria verna, Phellodendron amurense, and Reynoutria spp. have become invasive and pose a serious threat to remaining populations of native plants. Recent efforts to control the spread of invasive species and re-establish populations of extirpated species in the Thain Family Forest indicate that some of the changes to the Garden’s spontaneous flora may be reversible (Schuler & Forrest, 2016).

Since 2008, the Garden has conducted ecological restoration—or the process of assisting the recovery of an ecosystem that has been degraded, damaged or destroyed—in the 50 acre, remnant old-growth forest (SER, 2004). The main goal has been to improve forest health through active management informed by research. In this context, a healthy forest is defined as an ecosystem that is dominated by naturally regenerating populations of native plants that respond to and recover from disturbances of all scales with minimal human intervention, provides suitable habitat to a diversity of organisms in a structurally diverse, mixed-aged woodland, and sustains natural processes that support biodiversity (Schuler & Forrest, 2016). Forest inventories conducted in 1985, 2002, 2006, and 2011 have quantitatively revealed the expansion of non-native plant species in the forest (Rudnicky & McDonnell, 1989; Schuler & Forrest, 2016). Out of the 75 invasive plant species that are listed on the New York State Prohibited and Regulated List (NYS DEC, 2014), there are 34 extant species: Acer platanoides, Acer pseudoplatanus, Alliaria petiolata, Ampelopsis brevipedunculata, Anthriscus sylvestris, Aralia elata, Artemisia vulgaris, Celastrus orbiculatus, Cirsium arvense, Clematis terniflora, Euonymus alatus, Euonymus fortunei, Ficaria verna, Frangula alnus, Humulus japonicus, Lonicera japonica, Lonicera maackii, Lonicera morrowii, Lythrum salicaria, Microstegium vimineum, Miscanthus sinensis, Persicaria perfoliata, Phellodendron amurense, Potamogeton crispus, Reynoutria × bohemica, Reynoutria japonica, Reynoutria sachalinensis, Rhamnus cathartica, Robinia pseudoacacia, Rosa multiflora, Rubus phoenicolasius, Silphium perfoliatum, Vincetoxicum nigrum, and Vincetoxicum rossicum. It should be clarified that some of these plants are native to North America, just not to this region, e.g., Robinia pseudoacacia and Silphium perfoliatum. In addition, Berberis thunbergii and Phragmites australis are known to grow on the grounds of the Garden but, have not been vouchered, and Arthraxon hispidus, Euphorbia cyparissias, and Lespedeza cuneata are reported as historic but have not been collected recently. Interestingly, species of concern statewide and regionally, like Berberis thunbergii, Euonymus alatus, and Rhamnus cathartica are present only in limited numbers. Only half of the statewide list are currently extant or known to have been part of the Garden’s flora, leaving speculation for the potential of additional invasion in the future or possibly the habitat requirements for the remaining listed species do not exist in the Garden’s current 250 acres.

Invasive species of concern actively managed in 2008–2015 were Acer platanoides, Ailanthus altissima, Alliaria petiolata, Ampelopsis brevipedunculata, Aralia elata, Celastrus orbiculatus, Corydalis incisa, Ficaria verna, Hedera helix, Lonicera japonica, Lonicera maackii, Microstegium vimineum, Phellodendron amurense, Persicaria perfoliata, Reynoutria × bohemica, Reynoutria japonica, Reynoutria sachalinensis, Rosa multiflora, and Viburnum dilatatum. Extant non-native species that are present in inventory data but are not actively managed include Acer pseudoplatanus, Humulus japonicus, Paulownia tomentosa, Prunus avium, Prunus incisa, Prunus sargentii, Prunus subhirtella, Prunus × yedoensis, Rhamnus frangula, and Rubus phoenicolasius. Forest inventory data identify priority species and target locations for management. Literature is consulted for the best management practices for individual species with special focus on techniques that are proven effective at controlling a given species without causing unintended negative impacts on forest ecosystem processes (Schuler & Forrest, 2016). Published best management practices do not always result in the best control. Frequent monitoring and repeating the forest inventory sampling every five years reveals efficacy of control treatments (Schuler & Forrest, 2016). In addition, there is the constant threat of introduction of new invasive species as in the case in recent years of Corydalis incisa and Persicaria perfoliata. In both cases, the infestations were discovered early and mechanical management has been effective. Microstegium vimineum is abundant regionally, is observed periodically (despite treatment) along the Forest trails, and requires constant monitoring and management to prevent its spread. The monitoring for new invasive species and new infestation of extant species is constant.

When invasive plants are removed, it is necessary to plant desirable plants in management areas to prevent the reestablishment of the species removed or the establishment of another invasive species. From 2008 through 2015, Garden staff and volunteers spent a total of 21,500 hours in invasive plant management and restoration planting. Collectively, they have planted 25,220 individual plants plus seed mix including 11,700 herbaceous plants, 1720 shrubs, and 11,800 trees. Over 100 different species have been planted.

Restoring the historic flora and expanding the populations of rare extant species in the Forest has been an objective for ongoing restoration work. This flora has provided a list of historic or rare extant species of concern to be considered in future restoration projects. To date, the following historic or regionally rare species have been planted from regionally sourced seed: Anemone canadensis, Anemone nobilis, Anemone virginica, Anemone quinquefolia, Athryium filix-femina, Caltha palustris, Carex grayii, Carex plantaginea, Carex platyphylla, Corylus americana, Dryopteris marginalis, Elymus hystrix, Geranium maculatum, Geum canadensis, Iris versicolor, Lilium superbum, Lobelia siphilitica, Monarda fistulosa, Osmunda regalis, Osmundastrum cinnamomeum, Polystichum acrostichoides, Salix discolor, and Thelypteris palustris. Carex aggregata, Carex amphibola, Carex cephalaphora, and Carex debilis were vegetatively propagated from populations on the grounds of The New York Botanical Garden.