2008 NORTHEASTERN NATURALIST 15(2):159–176 Native and Introduced Jewelweeds of the Northeast Nava M. Tabak1,* and Eric von Wettberg2 Abstract - There are at least eight species of Impatiens (Jewelweeds) in the northeast, including native and non-native species, species of temperate and subtropical origins, and wild and planted species. The two native jewelweeds, I. capensis and I. pallida, are common in the region, and I. capensis has been introduced in Europe. Impatiens glandulifera is a non-native species that is locally well established and has the potential to become invasive in the region. The introduced I. balfourii and I. parviflora infrequently naturalize in the Northeast. Three Impatiens species of tropical origins, I. balsamina, I. walleriana, and I. hawkeri, may be able to establish in the Northeast with continued horticultural pressure. We review the descriptions and known distributions of these species, the direct experimental comparisons of their biological traits, and the histories of their introductions and horticultural uses. Such information is beneficial for assessing the invasive potential of these species in the northeast and abroad, and may help to set priorities for control efforts and regulation of horticultural uses. Introduction The genus Impatiens (Balsaminaceae) is globally distributed, and consists of about 850 to 1000 species (Clifton 2001, Mabberley 1997). Most Impatiens species are tropical or subtropical herbs and shrubs (Grey- Wilson, 1980a, b; Janssens et al. 2006; Yuan et al. 2004). The relatively few temperate species are annuals that grow in wet or moist habitats, and are commonly called jewelweeds or touch-me-nots. In the Northeast, there are two native jewelweeds, Impatiens capensis Meerb. (spotted touch-me not, orange jewelweed) and Impatiens pallida Nutt. (pale touch-me-not, yellow jewelweed). Many people recognize these plants by their distinctive flowers, and some use their clear sap as a poison ivy remedy. Impatiens glandulifera Royle (ornamental jewelweed, Himalayan balsam, policeman’s helmet) is a locally naturalized and potentially invasive introduced temperate species. Impatiens balfourii Hook. f. (Balfour’s touch-me-not) and I. parviflora DC. (smallflower touch-me-not) are also introduced temperate species, but with fewer instances of naturalization in the region. Impatiens balsamina L. (common garden balsam, spotted snapweed), I. walleriana Hook. f. (common African impatiens, Sultan’s impatiens, buzzy lizzy), and I. hawkeri W. Bull. (New Guinea Impatiens) are introduced species of tropical origins that have been documented as naturalized or capable of volunteering (i.e., growing in a site where previously cultivated from seeds released during cultivation). 1Hudsonia Ltd., PO Box 66, Red Hook, NY 12571. 2Section of Ecology and Evolution, University of California at Davis, Davis, CA 95616. *Corresponding author - tabak@bard.edu. 160 Northeastern Naturalist Vol. 15, No. 2 Although their common name implies a tendency to grow as weeds, temperate jewelweeds are seldom regarded as such in their native habitats. However, members of the Impatiens genus have not been exempt from the increase in trans-continental introductions and subsequent naturalizations and invasions of plants during the recent century. Indeed, the attractive flowers of some species of Impatiens make them popular ornamentals, and have lead to their intentional introduction. Their ability to naturalize (or form persistent populations without cultivation) and to become invasive in new ranges varies across the genus. Non-native plant invasions have been shown to have dramatic impacts on native ecosystems in many cases, and the science of invasive species is continuously expanding and leading to new insights on such impacts. Some Impatiens species have already become invasive in other parts of the world. Impatiens glandulifera, for example, has been the subject of numerous research and management projects in Europe, where it is considered to be one of the most invasive riparian weeds. The potential for members of this genus to become invasive in northeastern North America must be examined in order to prevent future invasions and to effectively manage existing populations. Understanding variability in potential for invasive behavior in the genus Impatiens is especially important because of the garden industry’s continuous marketing and distribution of new species of Impatiens, and the frequency with which certain Impatiens species are planted as bedding plants and other ornamentals. For example, sales of Impatiens walleriana (the most popular ornamental Impatiens in North America) are worth over 100 million US dollars a year (Jerardo 2005). Other species have grown in popularity in recent years (Jerardo 2005), and increased ornamental planting increases the risk of new invasions. Other factors, such as changes in patterns of land use and global climate change, are also likely to influence the risks of new invasions. Such risks can best be predicted and controlled by understanding the distributions and patterns of invasiveness in the genus. Herein, we provide a review of the Impatiens species that are native, naturalized, or likely to naturalize in the Northeast. This information is intended to assist in their identification and to inform future decisions about their management. Methods We reviewed the published literature on temperate Impatiens species to summarize their distributions and history of introductions to new ranges, with special emphasis on the northeastern United States (and the New England states in particular). We also reviewed the classification of some Impatiens species as invasive in their introduced ranges. Where available, we summarized research findings from direct experimental comparisons that have been made among temperate Impatiens species. To elucidate the historical patterns of ornamental planting of Impatiens species, we examined seed catalogues at the National Agricultural Library (NAL) in Washington, 2008 N.M. Tabak and E. von Wettberg 161 DC, following the protocols developed for other invasive plants of ornamental origin (Mack 1991). Results and Discussion Species profiles The Impatiens species present in the Northeast can be divided into those that have temperate native ranges (I. capensis, I. pallida, I. glandulifera, I. balfourii, and I. parviflora), and those that have tropical or subtropical native ranges (I. balsamina, I. walleriana, and I. hawkeri). In the Northeast, the temperate species tend to be taller than tropical and subtropical escapes, and often grow in self-thinning stands. Although in their native ranges some tropical species may grow to similar heights as some of the temperate species, their introductions in our range are usually the result of the escape of dwarfed cultivated varieties. Another general difference between these two groups is in flower shape: the temperate jewelweeds have somewhat elongated flowers due to the modified, saccate sepal with an elongated spur, while most tropical species that are common in cultivation have broader flowers and shorter saccate sepals and spurs (Caris et al. 2006; Grey-Wilson 1980a, b; Janssens et al. 2006). At least some physiological attributes of Impatiens species (e.g., seed dormancy) are also well correlated with their geographical distributions (Jouret 1977). The native Impatiens species are I. capensis and I. pallida. These two species are similar vegetatively, both growing to 1.5 m or more in height, and having glaucous or pale leaves that are ovate or elliptic and crenate. These two species can be differentiated by their flowers: I. capensis flowers are usually orange, and have a longer spur that curves sharply to parallel the flower, while I. pallida flowers are yellow and have blunt spurs that are bent at a 90° angle (Fernald 1950, corrected printing 1970; Gleason and Cronquist 1991) (Fig. 1). The shape of the spur is important for identification because there is well-documented variation in the color of I. capensis flowers, with two currently recognized forms: forma albiflora, with paler, creamy-white flowers, and forma immaculata, with spotless orange flowers (Weatherby 1917, Zika 2006b). An unusual characteristic of these two species of Impatiens is the presence of a mixed mating system, with cleistogamous (self-crossing and closed) and chasmogamous (outcrossing) flowers on the same individuals. The ability to self-pollinate may insure reproduction in small, marginal populations or under stressful conditions (Schemske 1978, Waller 1979). Impatiens capensis is widely distributed in moist habitats in the northeastern United States. Its native range includes all eastern states, most Canadian provinces, many mid-western states, and the Pacific Northwest (USDA 2007), although some of its current range in the Pacific Northwest may be a result of a more recent range extension (Ornduff 1967). Impatiens pallida, which is found locally in most eastern to mid-western states and eastern Canadian provinces (USDA 2007), has a more limited distribution 162 Northeastern Naturalist Vol. 15, No. 2 than I. capensis in the northeast. Despite frequently co-occurring, I. capensis and I. pallida do not appear to hybridize, and experimental pollen crosses do not set seed (Randall and Hilu 1990; E. von Wettberg, unpubl. data). Little information is available on species boundaries between other temperate Impatiens species (but see Ornduff 1967, Zika 2006a). Impatiens capensis, the more widely distributed of the two native jewelweeds, is also naturalized in Britain, France, central Europe, and Scandinavia (Krogerus 1977, Moore 1968, Pysek and Prach 1993). In England, I. capensis was first introduced as an ornamental to Surrey (south of London) in 1822, and it subsequently spread throughout the Thames watershed. It was secondarily introduced into the Norfolk broads tidal wetland system in the 1920s, and has spread widely in Norfolk and Suffolk (Preston et al. 2002). Despite its initial ornamental introduction, we are aware of no records of its continued cultivation or the commercial sale of its seeds in Europe; seeds can currently be purchased over the internet in North America, but viability Figure 1. Flowers of Impatiens spp. currently found in the Northeastern United States (clockwise from top left): I. capensis, I. pallida, I. glandulifera, and. I. balfourii. Photographs by Nava Tabak. 2008 N.M. Tabak and E. von Wettberg 163 is low due to drying out (Auyeung 2005, Dave’s Garden 2007). A comparison of the records of I. capensis with other introduced Impatiens species in English vice-counties (I. glandulifera and I. parviflora, see below) shows that I. capensis has the slowest rate of spread of these species, and occupies a smaller number of vice-counties (Perrins et al. 1993, Preston et al. 2002). Impatiens capensis does not occupy areas with high canopy cover (inferred by measuring light transmission) as frequently in its exotic range in England as it does in its native range in southeastern New England (E. von Wettberg, unpubl. data ). Because it spreads relatively slowly and does not form monocultures (Beerling and Perrins 1993, Preston et al. 2002), it is generally not considered to be a problematic weed in its introduced range (R. Ellis and R. Burton, Royal Botanical Survey, Norwich and London, UK, pers. comm.). Although Impatiens capensis does not hybridize with I. pallida, it does pose the threat of hybridization with I. noli-tangere L. (western touch-menot), a circumboreal species native to Europe (as well as northern Asia and probably Alaska), which has experienced declines in population numbers and sizes in recent years (Preston et al. 2002). It has been speculated that the two are in fact the same species (Williamson 1996), although they rarely, if ever, co-occur in Europe or Alaska (Preston et al. 2002), and there is no evidence of hybridization between these two species to date. Still, there is good reason to monitor for the possible hybridization of these two species. In Europe, I. capensis occupies habitats similar to those occupied by I. noli-tangere. They are close phylogenetic relatives, as all North American Impatiens are thought to descend from I. noli-tangere or a closely related taxon that crossed Beringia (Janssens et al. 2006, Yuan et al. 2004). Although the two Impatiens species native to the northeast do not readily hybridize with each other, I. capensis hybridizes with at least one other Impatiens species in the Pacific Northwest, and tropical Impatiens are often able to hybridize with each other (Ornduff 1967, Randall and Hilu 1990, Zika 2006a). The most notable introduction of the genus in the Northeast has been that of Impatiens glandulifera. This species is capable of growing to 3 m tall, has serrate leaves, nectar-tipped glands at its nodes and along the petioles, and purple, pink, or white flowers (Fig. 1). It is the only one of the jewelweeds in the Northeast with opposite or whorled leaves (Fig. 2). Impatiens glandulifera is native to the western Himalayas, where it grows in forests up to timberline, as well as more open and disturbed habitats (Gupta 1983, Pysek and Prach 1995). It was likely introduced into the United States from Europe, where it first naturalized in the United Kingdom in 1855 (Britten, as cited in Beerling and Perrins 1993). It has also been introduced to New Zealand (Weber 2003). The first record of its introduction in the northeastern United States is from a garden in Norwich, CT in 1883 (Mehrhoff et al. 2003), and it was cultivated on the grounds of the New York Botanical Garden in the early 1900s. It was also recorded from southern Vermont and Boston in the early 1900s, and subsequently from Washington and Knox counties in Maine (Mehrhoff et al. 2003, University of Maine Herbaria 2006). In New England, 164 Northeastern Naturalist Vol. 15, No. 2 Impatiens glandulifera is currently most widespread along the Maine coast, and is found in several small populations in the Housatonic River valley of Massachusetts, forming persistent populations in both regions. Isolated occurrences are also known from central Connecticut, north central Vermont, and inland localities in Maine (Fig. 3). It is reported from Ontario, Quebec, and the Canadian Maritime Provinces (Fernald 1950, corrected printing 1970; Mac- Quarrie and Schaefer 2001; Williamson 1996). Historic herbarium records from near Boston, MA, southern Vermont, and New Hampshire (Mehrhoff et al. 2003; Peter Zika, University of Washington Herbarium [WTU], Seattle, WA, pers. comm.) are difficult to confirm due to poor locality information. Impatiens glandulifera is also naturalized in several counties in New York State (Weldy and Werier 2005), and in the states of Michigan, Idaho, Montana, Figure 2. Habit of the top nodes of I m p a t i e n s glandulifera with flowers and immature fruit. 2008 N.M. Tabak and E. von Wettberg 165 Washington, Oregon, California (USDA 2007), and Alaska. This species is sold commercially, but it is not considered a popular ornamental in the Northeast. Anecdotal evidence of its introduction and spread in the region points to relatively few introductions for horticultural purposes, followed by dispersal from established populations by both natural and human-aided means (such as seed collecting, or picking flowering stems for bouquets and discarding them outside while they have viable seeds). Impatiens glandulifera is found in many northern and central European countries, where it is considered to be a serious weed of riparian zones, wetlands, woodlands, and waste grounds (Beerling and Perrins 1993, Dawson and Holland 1999, Drescher and Prots 2003, Environment Agency 2006, Grime et al. 1989, Pysek and Prach 1995, Scherer-Lorenzen et al. 2000, Weber 2003). As such, it has been well studied and targeted in weedmanagement plans in that region. Its invasiveness has been attributed to its great height, early synchronous germination followed by a high vegetative growth rate, relative frost tolerance, high rates of seed production, and Figure 3. The currently known distribution of Impatiens glandulifera in the New England states, as reconstructed from herbarium records and field observations. An herbarium record from New Hampshire that lacks other locality information is not shown. 166 Northeastern Naturalist Vol. 15, No. 2 other reproductive traits such as seed mass variation (Beerling and Perrins 1993, Grime 1979, Keonies and Glavac 1979, Perrins et al. 1990, Willis and Hulme 2004). Impatiens glandulifera has been shown to compete with native vegetation and possibly lead to greater erosion of riverbanks when it replaces perennial vegetation in Europe (Trautmann and Lohmeyer, as cited in Beerling and Perrins 1993; Cid-Benevento and Schaal 1986; Hulme and Bremner 2006; Keonies and Glavac 1979; Roblin 1994; Scherer-Lorenzen et al. 2000; Tickner et al. 2001a). In addition, the rich and abundant nectar produced by its flowers attracts pollinators at the expense of the pollination of co-occurring species in Europe (Chittka and Schurkens 2001), and it is unlikely that its success will be limited by pollinators in the Northeast (Tabak 2005). However, recent studies in Europe show that, at least in the short term, the effects of I. glandulifera invasions on plant communities may not be as severe as those of other invasive riparian weeds, and its removal in disturbed riparian habitats is likely to encourage colonization by other nonnative weeds (Hejda and Pysek 2006, Hulme and Bremner 2006). Another introduction from the Himalayas has been that of Impatiens balfourii (Fig. 1). This species is sometimes confused with I. glandulifera because its leaves are serrate and the flowers are white to purple. However, Impatiens balfourii is smaller in stature than I. glandulifera (reaching a maximum height of 0.8 m [Moore 1968]), has alternate leaves, and its flower spurs are long and straight as opposed to I. glandulifera’s blunt and curved flower spurs. In contrast to I. glandulifera, this temperate jewelweed is only casually naturalized in the Northeast, although it is established in other states (USDA 2007, Zika 2006a). It is also naturalized in disturbed habitats in central and southern Europe (Moore 1968), but is not considered to be invasive there. Impatiens balfourii’s naturalization and spread in the Northeast is probably limited most by its relatively low frost tolerance. Seeds of I. balfourii are now sold commercially for horticultural purposes, but it is not a common planting in northeastern gardens, and we did not find records of its sale before 1970 in our seed catalog search (E. von Wettberg, unpubl. data). Impatiens parviflora is a temperate central Asian species that grows to 1 m tall and has small yellow flowers with straight spurs. The earliest of these flowers are often cleistogamous (Moore 1968). It is widely naturalized in Europe and has been infrequently reported from Quebec, New Brunswick, Nova Scotia, and Prince Edward Island (Fernald 1950, corrected printing 1970; Gleason and Cronquist 1991; Kartesz and Meacham 1999; Kornas 1990; Williamson 1996). To our knowledge, this species has not yet been recorded as naturalized from any state in the northeastern United States. Although it occupied disturbed habitats when it first naturalized in Europe, it subsequently invaded undisturbed deciduous forests, giving it the status of an invasive species (Kornas 1990). Its ability to grow and reproduce in low light levels is considered the main factor enabling its spread into forests, but its regional spread may be limited by a low tolerance of frost (Hughes 2008 N.M. Tabak and E. von Wettberg 167 1965, Perrins et al. 1993). Although its naturalizations in Europe and North America were likely a consequence of escape from ornamental planting, its flowers are not showy, there are no records of its sale in catalogs held at the US National Agricultural Library before 1970 (E. von Wettberg, unpubl. data), and it is not commonly sold for ornamental purposes now. The three most widely planted ornamental species of Impatiens, I. balsamina, I. walleriana, and I. hawkeri, are of tropical or subtropical origins, and do not commonly naturalize in northeastern North America. Published information on their ecology in their native ranges is limited at this time. Of the three species, I. balsamina, has been in ornamental cultivation the longest in North America, with seeds having been commercially available since before 1870 and now being widely sold in of seed catalogues from multiple locations in the eastern US (Fig. 4). Impatiens balsamina is believed to be native to India, but because it has been in ornamental cultivation for a long time in Europe, its exact native range is difficult to discern. Although I. balsamina does occasionally escape from cultivation in the Northeast (Fernald 1950, corrected printing 1970), this appears to be mostly restricted to gardens and other sheltered locations. It spreads locally in a number of southern and central Atlantic states (USDA 2007). Impatiens walleriana is native to tropical East Africa (Grey-Wilson 1980b), and has become established in a large portion of the wet tropics, from Central and South America to Queensland (Australia), northern New Zealand, Hawaii and Florida, and potentially parts of Asia (Mandle et al., in press). While I. walleriana has been marketed in the US since around 1890 (Fig. 4), it became extremely popular as a bedding plant in the 1960s following the introduction of dwarf genotypes bred in Costa Rica by Claude Hope (Pollan 2001). It now constitutes a $120-million bedding-plant industry in the US, creating a propagule pressure of millions of plants across most of North America, including theNortheast (Jerardo 2005). Despite this considerable propagule pressure and the occasional occurrence of volunteers in gardens the year after planting, it does not appear to readily naturalize in the Northeast. Climatic modeling of this species’ native and exotic ranges suggests that it cannot establish in areas with freezing temperatures (Mandle et al., in press). This limitation is probably due to a lack of seed dormancy, which enables the temperate jewelweed species to delay germination until the spring. Impatiens hawkeri was introduced from Indonesia following a collecting trip made in 1970 by Robert Armstrong of Longwood Gardens, PA (Benjamin 1990). Although it has become a popular ornamental due to its tolerance of full sun, it is currently not planted as widely as I. walleriana (Jerardo 2005). Historical trends in its planting could not be discerned from the National Agricultural Library catalog holdings because catalogues are not held chronologically after 1960, making inference about its historical spread more difficult. Its tropical origin is likely to preclude it from naturalizing in temperate areas, but it can volunteer after ornamental planting. Because I. hawkeri 168 Northeastern Naturalist Vol. 15, No. 2 may be part of a species complex of other Indonesian and New Guinean Impatiens species, and is frequently crossed to develop new ornamental varieties (Benjamin 1990), it is possible that a cross with a temperate or tropical montane Impatiens will confer some cold tolerance on this ornamental, and allow full naturalization in temperate areas in the future. Comparisons of temperate Impatiens species The more limited distribution of Impatiens pallida as compared to I. capensis within the northeast region is generally attributed to its affinity for Figure 4. Number of Impatiens walleriana and I. balsamina vendors in the US, estimated from seed catalogues at the US National Agricultural Library (NAL) in Washington, DC. Following the protocols of Mack (1991), we sampled catalogues for all Impatiens species at 10-year intervals from 1870 until 1970, and all catalogues before 1870. After 1880, we sub-sampled within the alphabet because of the prohibitively large number of catalogues. We calculated estimated number of vendors by multiplying the number of vendors observed by the amount of the alphabet sampled. There is no geographic or alphabetical bias to this sample. Total seed catalogue estimates are low for all species after 1950 because the NAL does not store catalogues after 1960 by year, making examination of historical trends difficult for this time period. 2008 N.M. Tabak and E. von Wettberg 169 calcareous soils and shadier habitats (Fernald 1950, corrected printing 1970; Gleason and Cronquist 1991). However, a study in Quebec confirming that I. pallida is constrained to a narrower range of habitats found that these habitats had higher soil nitrate, potassium, and phosphorous, were drier, and had lower soil pH (Lechowicz et al. 1988). These findings somewhat contradict the observation-based general understanding of I. pallida’s restricted distribution, and more work is needed to clarify this point. Experimental comparisons of Impatiens glandulifera with I. capensis in New England show that the two species can grow in similar habitats, but I. glandulifera plants can also tolerate drier soil conditions and have higher relative growth rates than I. capensis plants across a full range of soilmoisture and light conditions (Tabak 2005). The common name touch-me-not comes from the explosive nature of the seed capsules of plants in this genus. Long-distance seed dispersal in the northeastern species can occur when seeds are shot out of the capsules and fall into the moving water of rivers or streams. The seeds of both I. glandulifera and I. capensis can float. Impatiens glandulifera seeds will float in still water under laboratory conditions for as long as forty days, while I. capensis seeds can float for at least 200 days (Auyeung 2005). Increased water salinity increases these seed floating times, but also lowers the germination rates of seeds. By modeling dispersal in rivers of varying speeds, these differences in floating ability translate into substantial differences in potential propagule pressure in downstream areas of watersheds invaded by either of these species: I. glandulifera invasions will be characterized by much higher local densities and more of a uniform front regardless of river speeds, while a large proportion of I. capensis seeds may be flushed from faster moving river systems to spread widely(Auyeung 2005). Impatiens pallida seeds are also known to float for long periods of time (Cid-Benevento and Schaal 1986), but the ability of seeds of other Impatiens species to float is still untested. Given their preference for wet habitats, floating seeds and water-aided dispersal are probably common within this genus. The seeds of Impatiens glandulifera, I. capensis, and I. pallida germinate almost synchronously in the early spring, after the winter’s cold temperatures break their physiological dormancy (Barton 1939, Baskin and Baskin 1998, Grime 1979, Jouret 1976, Leck 1979, Mumford 1988, Nozzolillo and Thie 1983, Winsor 1983). Impatiens balfourii and I. parviflora seeds also require a cold stratification to break their dormancy (Jouret 1976). Germination experiments show that I. glandulifera seeds are capable of high germination across a broader range of soil moisture than I. capensis seeds (Tabak 2005). Impatiens capensis and I. pallida seeds are known to be recalcitrant and dependant on being fully imbibed during cold stratification in order to come out of dormancy (Jouret 1976, Leck 1979, Nozzolillo and Thie 1983, Tabak 2005). In contrast, Impatiens glandulifera and I. balfourii seeds are able to germinate even after periods of drying (Jouret 1976, 1977; Kollmann and Banuelos 2004; Mumford 1988; Tabak 2005; Tickner et al. 2001b). Jouret 170 Northeastern Naturalist Vol. 15, No. 2 (1976, 1977) reported that in conjunction with cold stratification, seed desiccation is a primary dormancy-breaking mechanism for these two species in their native range. Desiccation is a requirement for breaking the dormancy of Impatiens balsamina seeds (Jouret 1976). In a common garden experiment, Impatiens glandulifera seedlings had better establishment rates and a greater tolerance to frost than I. parviflora and I. balfourii (Perrins et al. 1993). Impatiens glandulifera also has a higher number of seeds per capsule than I. balfourii, I. parviflora, I. capensis, and I. pallida (Perrins et al. 1993; N.M. Tabak, unpubl. data). The number of capsules per plant can vary greatly in these species depending on plant size and population densities, but the maximum seed production rates per plant are greater for I. glandulifera than for the two native Impatiens species (Beerling and Perrins 1993, Cid-Benevento and Schaal 1986, Lechowicz et al. 1988). Impatiens parviflora is more shade tolerant than I. glandulifera, I. balfourii, and I. capensis (Hughes 1965, Jouret 1977, Lee 1984, Perrins et al. 1993). Discussion and recommendations The introduced Impatiens species should be scrutinized for their potential impacts in the northeast. Those species that have a history of invasiveness in other temperate regions are of particular concern (Reichard and Hamilton 1997). The size of the native ranges of the temperate Impatiens is not positively correlated with their invasive abilities, as is the case with some other plants (Williamson 1996), and should not be used to predict invasiveness. Jouret (1977) described the strong adaptive correlations between the seed dormancies and the distributions of some Impatiens species, and other traits should also reflect adaptations to the conditions in these plants’ native ranges. As can be expected from such a biogeographical perspective, most of the naturalized or invasive Impatiens species in temperate regions are those of temperate origins and not the more widely planted tropical species. Of the introduced species of Impatiens, I. glandulifera presents the clearest threat of becoming invasive in the region, and indeed can already be considered invasive in the Northeast on a local scale. It has been very well studied in Europe where it is considered to be highly invasive, and recent research suggests that it has a similar potential for biological spread in the New England states (Tabak 2005). Its less aggressive spread throughout the Northeast in comparison with parts of Europe may reflect a longer lag phase due to comparatively lighter propagule pressure, fewer human-aided dispersal opportunities, or less-frequent disturbance-driven establishment opportunities. The ability of I. glandulifera seeds to maintain their viability in a dry state provides the species with added dispersal opportunities as compared with the native species: humans, whether individuals or seed companies, can collect the seeds and re-distribute them without maintaining them in moist conditions. Preventive restrictions on the sale of I. glandulifera in the Northeast are well warranted in light of its current distribution and behavior in the Northeast. Impatiens parviflora, which is naturalized in eastern Canadian provinces, should 2008 N.M. Tabak and E. von Wettberg 171 be studied and observed. Considering its ability to naturalize and invade woodlands in Europe and its naturalization in Canada, its arrival in northern New England should be anticipated. Although not widely distributed in the Northeast, I. balfourii has also shown the ability to naturalize in more than one new region and deserves careful observation. Even though the body of research on the invasive potential of I. balfourii and I. parviflora is still lacking, restrictions on the sale of these potentially invasive species would be a logical precaution while further research is conducted on their potential biological and ecological limitations in the region. The tropical and temperate groups of Impatiens are generally kept distinct by their geographical ranges. Tropical and subtropical species carry a low risk of naturalization in temperate regions due to a lack of tolerance for cold temperatures, while temperate species carry a low risk of naturalization in tropical regions because of adaptations such as their seeds’ requirement of cold stratification. This pattern provides a useful general guideline for safe ornamental uses of Impatiens species. However, global climate change is expected to have great impacts on the future distributions of plants, and must be considered as part of any analysis of potential range expansions. Increasing temperatures allow for the possibility that subtropical species that are not currently problematic will become so in the future because of reduced occurrences of killing frosts and elevated winter temperatures that allow for greater seed survival. Impatiens balsamina is already established in many temperate areas, and warming climate conditions may facilitate its further spread. The other widely planted tropical species may also become naturalized with continued planting and regional warming, as their exotic ranges already include areas with cooler climates than their native ranges (Mandle et al., in press). Additional information on the native ranges and habitat requirements of introduced Impatiens species of tropical origins will improve our understanding of their ability to extend their ranges. Global climate change will also impact the ranges of temperate Impatiens species whose current ranges are limited by temperature. Predictions of changes in the range of I. glandulifera in Europe in response to warming temperatures include range expansions in a northerly direction as well as colonization of higher altitudes (Beerling 1993, Beerling and Woodward 1994). Although predictive tools are not yet considered to be perfected, such a pattern may reasonably be expected for other temperate Impatiens species whose northern limits are constrained by climate in northeastern North America. Changes in southern range limits for these species are also likely, but have not yet been closely examined. Species boundaries in the genus Impatiens require further investigation, as hybridization may also facilitate new naturalizations. Additional work is needed to clarify the potential for hybridization both between and among the temperate and subtropical species, and their possible consequences. The hybridization of Impatiens capensis with a temperate congener in the Pacific Northwest (Ornduff 1967, Zika 2006a) is 172 Northeastern Naturalist Vol. 15, No. 2 of concern considering its wide distribution in the Northeast and range overlap with introduced temperate Impatiens species. Hybridization with a temperate species may provide tropical ornamental Impatiens species with the genes necessary to naturalize and even become invasive. Thus, a better understanding of the ability of I. capensis and other temperate species to hybridize is essential. The traits bred into domesticated tropical Impatiens, such as dwarfism and various flower colors, may have impacts on competitive ability and relationships with pollinators in new ranges. A research focus on those species originating in areas where the boundaries within the genus remain unclear, such as East Africa, Madagascar, and southeast Asia-Indonesia, will be especially informative. Multiple introductions from these regions have the potential to form novel hybrids with invasive traits. Of particular concern are tropical montane forms, which may have an increased tolerance of temperate conditions. Examination of hybridization potential should precede the introduction of new ornamental species of Impatiens, so as to guard against the formation of hybrid invaders and the genetic erosion of native species. Additional populationlevel molecular systematic work will help to clarify the relationships of many putative taxa in the Pacific Northwest, providing useful information on species boundaries within the genus (Ornduff 1967, Zika 2006a). Acknowledgments We thank Christopher Mattrick, Lisa Mandle, Andy Reinmann, and two anonymous reviewers for comments on this manuscript. We are grateful to Les Mehrhoff, John Silander, and Johanna Schmitt for their advice on invasion biology and Impatiens. Rodney Burton and Bob Ellis of the British Botanical Survey and Mark Williamson provided insights into Impatiens in the UK and assistance in finding I. capensis populations in the UK. Richard Mack and Lisa Mandle helped with our seed catalogue search. Stephane Marty and Andrew Latimer provided translation services. We are grateful to the many people who helped update the distribution of I. glandulifera in New England and other Impatiens species in the Northeast. This research was supported by the Invasive Plant Atlas of New England (IPANE, USDA CSREES grant), Ronald Bamford Endowment grant to N.M. Tabak, NSF DDIG 0408015 to E. von Wettberg, a SICB graduate student travel grant to E. von Wettberg, a Sigma Xi grant in aid of research to E. von Wettberg, and an EPA STAR graduate fellowship to E. von Wettberg. Literature Cited Auyeung, D. 2005. Desiccation tolerance and seed floating ability of escaped ornamental Impatiens. Center for Environmental Studies. Brown University, Providence, RI. Barton, L.V. 1939. Experiments at Boyce Thompson Institute on germination and dormancy in seeds. Scientific Horticulture 7:186–193. Baskin, C.C., and J.M. Baskin 1998. Seeds: Ecology, Biogeography, and Evolution of Dormancy and Germination. Academic Press, San Diego, CA. Beerling, D.J. 1993. The impact of temperature on the northern distribution limits of the introduced species Fallopia japonica and Impatiens glandulifera in northwest Europe. Journal of Biogeography 20:45–53. 2008 N.M. Tabak and E. von Wettberg 173 Beerling, D.J., and J.M. Perrins. 1993. Impatiens glandulifera Royle (Impatiens roylei Walp). Journal of Ecology 81:367–382. Beerling, D.J., and F.I. Woodward. 1994. Climate change and the British scene. Journal of Ecology 82:391–397. Benjamin, J.M. 1990. History and development of New Guinea Impatiens. Horticulture. University of Delaware, Dover, DE. Caris, P.L., K.P. Geuten, S.B. Janssens, and E.F. Smets. 2006. Floral development in three species of Impatiens (Balsaminaceae). American Journal of Botany 93:1–14. Chittka, L., and S. Schurkens. 2001. Successful invasion of a floral market. Nature 411:653–653. Cid-Benevento, C.R., and B.A. Schaal. 1986. Variation in population growth rate in the woodland annual Impatiens pallida (Balsaminaceae). American Journal of Botany 73:1031–1042. Clifton, R. 2001. Geraniales Species Checklist Series: Volume 6. Geraniaceae Bentham and Hooker, tribe VII Balsaminaceae. The Geraniaceae Group, Dover, UK. Dave’s Garden. 2007. Spotted jewelweed, touch-me-not, Impatiens capensis. Dave’s Garden. Website. Available online at http://davesgarden.com/pf/go/1431/index. html. Accessed January 22, 2007. Dawson, F.H., and D. Holland. 1999. The distribution in bankside habitats of three alien invasive plants in the UK in relation to the development of control strategies. Hydrobiologia 415:193–201. Drescher, A., and B. Prots. 2003. Distribution patterns of the Himalayan balsam (Impatiens glandulifera Royle) in Austria. Pp. 137–146, In A. Zajac, M. Zajac, and B. Zemanek (Eds.). Phytogeographical Problems of Synanthropic Plants. Institute of Botany, Jagiellonian University, Cracow, Poland. Environment Agency. 2006. Top-ten most-wanted foreign species. Head Office Press Office. London, UK. Fernald, M.L. 1950, corrected printing 1970. Gray’s Manual of Botany. Dioscordies Press, Portland, OR. Gleason, H.A., and A. Cronquist. 1991. Manual of Vascular Plants of Northeastern United States and Adjacent Canada. New York Botanical Garden, New York, NY. Grey-Wilson, C. 1980a. Impatiens in Pauasia. Studies in Balsaminaceae I. Kew Bulletin 34:661–688. Grey-Wilson, C. 1980b. Impatiens of Africa. A.A. Balkema, Rotterdam, The Netherlands. Grime, J.P. 1979. Plant Strategies and Vegetation Processes. Wiley, Chichester, UK. Grime, J.P., J.G. Hodgson, and R. Hunt. 1989. Comparative Plant Ecology: A Functional Approach to Common British Species. Chapman and Hall, London, UK. Gupta, R.K. 1983. The Living Himalayas, Volume I: Aspects of Environmental and Resources Ecology of Garhwal. Today and Tomorrow’s Printers and Publishers, New Delhi, India. Hejda, M., and P. Pysek. 2006. What is the impact of Impatiens glandulifera on species diversity of invaded riparian vegetation? Biological Conservation 132:143–152. Hughes, A.P. 1965. Plant growth and the aerial environment VII. Growth of Impatiens parviflora in very low light intensities. New Phytologist 64:55–64. 174 Northeastern Naturalist Vol. 15, No. 2 Hulme, P.E., and E.T. Bremner. 2006. Assessing the impact of Impatiens glandulifera on riparian habitats: Partitioning diversity components following species removal. Journal of Applied Ecology 43:43–50. Janssens, S., K. Geuten, Y.M. Yuan, Y. Song, P. Kupfer, and E. Smets. 2006. Phylogenetics of Impatiens and Hydrocera (Balsaminaceae) using chloroplast atpBrbcL Spacer sequences. Systematic Botany 31:171–180. Jerardo, A. 2005. Floriculture and nursery crops situation and outlook yearbook. Economic Research Service. USDA, Beltsville, MD. Jouret, M.F. 1976. Ecology of seed dormancy and germination for various species of the genus Impatiens. Bulletin de la Societé de Botanique de Belgique 103:213–225. Jouret, M.F. 1977. Relationship between seed dormancy and germination for various species of the genus Impatiens. Bulletin de la Society de Botanique de Belgique 110:119–124. Kartesz, J.T., and C.A. Meacham 1999. Synthesis of the North American Flora. North Carolina Botanical Garden, Chapel Hill, NC. Keonies, H., and V. Glavac. 1979. Über die Konkurrenzfähigkeit des Indischen Springkrautes (Impatiens glandulifera Royle) am Fuldaufer bei Kassel. Phillipia 44:47–59. Kollmann, J., and M.J. Banuelos. 2004. Latitudinal trends in growth and phenology of the invasive alien plant Impatiens glandulifera (Balsaminaceae). Diversity and Distributions 10:377–385. Kornas, J. 1990. Plant invasions in central Europe: Historical and ecological aspects. Pp. 19–36, In F. diCastri, A.J. Hansen, and M. Debussche (Eds.). Biological Invasions in Europe and the Mediterranean Basin. Kluwer Academic Publishers, Dordrecht, The Netherlands. Krogerus, H. 1977. Impatiens capensis introduced into the Lojo-area in SW Finland. Memoranda Society Fauna Flora Fennica 53:99–100. Lechowicz, M.J., D.J. Schoen, and G. Bell. 1988. Environmental correlates of habitat distribution and fitness components in Impatiens capensis and Impatiens pallida. Journal of Ecology 76:1043–1054. Leck, M.A. 1979. Germination behavior of Impatiens capensis Meerb. (Balsaminaceae). Bartonia 46:1–11. Lee, J.C. 1984. The ecology of Impatiens glandulifera. A study of an alien species. Unpublished B.Sc. Project. York University, York, UK. Mabberley, D.J. 1997. The Plant Book. Cambridge University Press, Cambridge, UK. Mack, R.N. 1991. The commercial seed trade: An early disperser of weeds in the United States. Economic Botany 45:257–273. MacQuarrie, K., and H. Schaefer. 2001. New plant records for Prince Edward Island. Canadian Field-Naturalist 115:446–450. Mandle, L.A., E.J. von Wettberg, M. Hoffman, and J. Schmitt. In Press. Escape from a climate cage? Impatiens walleriana has a broader climate envelope outside its native range. Conservation Biology. Mehrhoff, L.J., J.A. Silander, S.A. Leicht, E.S. Mosher, and N.M. Tabak. 2003. IPANE: Invasive Plant Atlas of New England. Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT. Available online at http://ipane.org. Accessed January 2007. 2008 N.M. Tabak and E. von Wettberg 175 Moore, D.M. 1968. Impatiens. Pp. 240–241, In T.G. Tutin, V.H. Heywood, N.A. Burges, D.H. Valentine, S.M. Walters, and D.A. Webb (Eds.). Flora Europaea. Cambridge University Press, Cambridge, UK. Mumford, P.M. 1988. Alleviation and induction of dormancy by temperature in Impatiens glandulifera Royle. New Phytologist 109:107–110. Nozzolillo, C., and I. Thie. 1983. Aspects of germination of Impatiens capensis Meerb., formae capensis and immaculata and I. pallida Nutt. Bulletin of the Torrey Botanical Club 110:335–344. Ornduff, R. 1967. Hybridization and regional variation in Pacific Northwestern Impatiens (Balsaminaceae). Brittonia 19:122–128. Perrins, J., A. Fitter, and M. Williamson. 1990. What makes Impatiens glandulifera invasive? BES Industrial Ecology Group Conference, Cardiff, UK. Perrins, J., A. Fitter, and M. Williamson. 1993. Population biology and rates of invasion of three introduced Impatiens species in the British Isles. Journal of Biogeography 20:33–44. Pollan, M. 2001. Floral conquest, an obituary for Claude Hope. Page SM 26. New York Times, New York, NY. Preston, C.D., D.A. Pearman, and T.D. Dines. 2002. New Atlas of the British and Irish Flora: An Atlas of the Vascular Plants of Britain, Ireland, the Isle of Man, and the Channel Islands. Oxford University Press, Oxford, UK. Pysek, P., and K. Prach. 1993. Plant invasions and the role of riparian habitats: A comparison of four species alien to central Europe. Journal of Biogeography 20:413–420. Pysek, P., and K. Prach. 1995. Invasion dynamics of Impatiens glandulifera: A century of spreading reconstructed. Biological Conservation 74:41–48. Randall, J.L., and K.W. Hilu. 1990. Interference through improper pollen transfer in mixed stands of Impatiens capensis and I. pallida (Balsaminaceae). American Journal of Botany 77:939–944. Reichard, S.H., and C.W. Hamilton. 1997. Predicting invasions of woody plants introduced into North America. Conservation Biology 11:193–203. Roblin, L. 1994. Alien invasive weeds: An example of the National Rivers Authority sponsored research. Pp. 189–193, In L.C. de Waal, L.E. Child, P.M. Wade, and J.H. Brock (Eds.). Ecology and Management of Invasive Riverside Plants. John Wiley and Sons, Ltd., Chichester, UK. Schemske, D.W. 1978. Evolution of reproductive characteristics in Impatiens (Balsaminaceae): The significance of cleistogamy and chasogamy. Ecology 59:596–613. Scherer-Lorenzen, M., A. Elend, S. Nollert, and E.D. Schulze. 2000. Plant Invasions in Germany: General Aspects and Impacts of Nitrogen Deposition. Island Press, Washington, DC. Tabak, N.M. 2005. Assessing the invasive potential of Impatiens glandulifera (ornamental jewelweed) in New England. Ecology and evolutionary biology. University of Connecticut, Storrs, CT. Tickner, D.P., P.G. Angold, A.M. Gurnell, and J.O. Mountford. 2001a. Riparian plant invasions: Hydrogeomorphological control and ecological impacts. Progress in Physical Geography 25:22–52. 176 Northeastern Naturalist Vol. 15, No. 2 Tickner, D.P., P.G. Angold, A.M. Gurnell, J.O. Mountford, and T. Sparks. 2001b. Hydrology as an influence on invasion: Experimental investigations into competition between the alien Impatiens glandulifera and the native Urtica dioica in the UK. Pp. 159–168, In G. Brundu, J.L. Brock, I. Camarda, L.E. Child, and M. Wade (Eds.). Plant Invasions: Species Ecology and Ecosystem Management. Backhys Publishers, Leiden, The Netherlands. University of Maine Herbaria. 2006. University of Maine. Orono, ME. Website. Available at http://www.umesci.maine.edu/biology/herbarium. Accessed January 2007. US Department of Agriculture (USDA). 2007. Synthesis, United States Department of Agriculture Plants Database. NRCS USDA, Beltsville, MD. Available at http://plants.usda.gov. Accessed January 2007. Waller, D.M. 1979. The relative cost of self- and cross-fertilized seeds in Impatiens capensis (Balsaminaceae). American Journal of Botany 66:213–320. Weatherby, C.A. 1917. Color forms of Impatiens biflora. Rhodora 19:115–118. Weber, E. 2003. Invasive Plant Species of the World: A Reference Guide to Environmental Weeds. CABI Publishing, Wallingford, UK. Weldy, T., and D. Werier. 2005. New York Flora Atlas. (S.M. Landry, K.N. Campbell, and L.D. Mabe [original application development], Florida Center for Community Design and Research, University of South Florida). New York Flora Association, Albany, NY. Williamson, M. 1996. Biological Invasions. Chapman and Hall, London, UK. Willis, S.G., and P.E. Hulme. 2004. Environmental severity and variation in the reproductive traits of Impatiens glandulifera. Functional Ecology 18:887–898. Winsor, J. 1983. Persistence by habitat domination in the annual Impatiens capensis (Balsaminaceae). Journal of Ecology 71:451–466. Yuan, Y.M., Y. Song, K. Geuten, E. Rahelivololona, S. Wohlhauser, E. Fischer, E. Smets, and P. Kupfer. 2004. Phylogeny and biogeography of Balsaminaceae inferred from ITS sequences. Taxon 53:391–403. Zika, P.F. 2006a. Impatiens x pacifica (Balsaminaceae), a new hybrid jewelweed from the Pacific Northwest coast of North America. Novon 16:443–448. Zika, P.F. 2006b. Lectotypification of the names of two color forms of Impatiens capensis (Balsaminaceae). Rhodora 108:62–64.