The Asian needle ant, Brachyponera chinensis (Emery) (Hymenoptera: Formicidae), is an East Asian species currently spreading through the eastern US. Although not aggressive, B. chinensis has a painful sting that can induce a severe allergic reaction in humans and disrupt native ecological communities. To document the global distribution of B. chinensis and evaluate its potential for further spread, we compiled specimen records from over 780 sites in its native and introduced ranges. Unfortunately, in its native range, B. chinensis often has been confused with morphologically similar species, resulting in some unreliable published distribution records. Therefore, we designated confirmed identification records as “B. chinensis s.s.” (sensu stricto). Our report chronicles the earliest known records of B. chinensis for 50 geographic areas in the Old World (36 countries and island groups) and New World (17 states in the US) and the recent expansion of B. chinensis into the eastern coast of the Black Sea along with new state distribution records from the US. We also provide new information on the phenology of B. chinensis. This ant species represents a potentially serious public health threat that warrants more epidemiological study.
Many invasive ant species cause major economic losses globally (Bradshaw et al. 2016) and negatively impact ecosystem functions and the associated native flora and fauna (Holway et al. 2002). The biology and distribution of some species, such as the red imported fire ant (Solenopsis invicta Buren), Argentine ant (Linepithema humile [Mayr]), and little fire ant (Wasmannia auropunctata [Roger]) (all Hymenoptera: Formicidae) have been extensively studied for several decades (Holway et al. 2002). However, other ant invaders have been recognized only recently, and key aspects of the ecology and spread of these species remain unresolved.
The Asian needle ant, Brachyponera chinensis (Emery) (Formicidae: Brachyponera) (formerly Pachycondyla chinensis Emery) (Hymenoptera: Formicidae), is an East Asian ponerine (Ponerinae) currently spreading through the eastern US. Although not aggressive, this species has a painful sting that can cause severe allergic reactions in humans (Smith 1979; Green 1992; Xu 1994; Cho et al. 2002; Fukuzawa et al. 2002; Leath et al. 2006; Nelder et al. 2006; Zungoli et al. 2006). Brachyponera chinensis is widespread in East Asia and can be quite common in some regions. For instance, Brown (1958) wrote that in China, B. chinensis “is seen everywhere on the rice paddy dikes and in the farm compounds, foraging in the open on tree trunks and on the ground in broad daylight.” Smith (1934) first reported B. chinensis (as Euponera solitaria Smith) in the US, from sites in Georgia, North Carolina, and Virginia. More recently, several studies in North and South Carolina documented the largely overlooked threat that this ant poses (Nelder et al. 2006; Zungoli et al. 2006; Guénard & Dunn 2010). In the eastern US, B. chinensis has had a detrimental impact on native ants and other insects as well as on ant-seed dispersal mechanisms (Guénard & Dunn 2010; Rodriguez-Cabal et al. 2012; Warren et al. 2015; Suehiro et al. 2017; B. G. unpublished data). In contrast to most other invasive ants in the southeastern US, which are mostly located in disturbed habitats, B. chinensis commonly occurs in undisturbed forested habitats (Guénard & Dunn 2010; Warren et al. 2015; Suehiro et al. 2017). However, B. chinensis also has been collected in urban habitats with little vegetative cover (Menke et al. 2011; Guénard et al. 2014).
Unfortunately, identification of B. chinensis in its native range has been difficult due to the presence of other morphologically similar species, such as B. luteipes (Mayr), B. nigrita (Emery), and B. obscurans (Walker). Indeed, Brown (1958) recognized the taxonomic uncertainty of the B. chinensis complex, which he considered “taxonomically confused.” Brachyponera chinensis from Shanghai, China, was described by Emery (1895) as Ponera nigrita chinensis, but Smith (1874) had previously described the species collected from Japan as Ponera solitaria. The name solitaria was unavailable because it was already occupied by a senior homonym, Ponera solitaria Smith, 1860 (= Pachycondyla solitaria), described from Bacan, Indonesia. Brown (1958) thought that B. chinensis might be a junior synonym of Brachyponera luteipes, and Taylor (1961) and Wilson and Taylor (1967) listed Brown's (1958) record from New Zealand as B. luteipes. However, Yashiro et al. (2010) confirmed that B. chinensis, B. luteipes, and B. nigrita (Emery) were 3 distinct species, based on consistent morphological and DNA sequence differences. For example, B. chinensis is larger than B. luteipes, but smaller than B. nigrita (Yamane 2007). Through numerous revisions, different authors have placed B. chinensis in 4 different genera: Ponera, Euponera, Brachyponera, and Pachycondyla. Most recently, Schmidt and Shattuck (2014) removed B. chinensis from Pachycondyla, placing it in the newly revived genus Brachyponera. In addition, a new Japanese species that was syntopic with B. chinensis, Brachyponera nakasujii, was recently described and was distinguished from B. chinensis by genetic and morphological analyses (Yashiro et al. 2010). Thus, the identity of ants from earlier records of B. chinensis from Asia should be viewed with caution. For example, Yashiro et al. (2010) concluded that the “B. chinensis” ants reported by Gotoh and Ito (2008) were most likely B. nakasujii. For the purposes of this study, we referred to confirmed B. chinensis records as “B. chinensis s.s.” (sensu stricto), which include our own verified records of B. chinensis, those of Yashiro et al. (2010), the type specimen from Shanghai, China (Emery 1895), and recent records where the authors explicitly distinguished B. chinensis from B. nakasujii (e.g., Harada et al. 2012, 2014; Fukumoto & Yamane 2015). All North American records were B. chinensis s.s. (Yashiro et al. 2010).
Here, we analyze and synthesize the current known distribution of B. chinensis in its native and introduced ranges, considering the taxonomic uncertainty and the timing of invasion in North America. Based on these results, we discuss the invasion potential of this ant. We also present some elements of its reproductive biology and related data addressing the potential threat that this species represents for humans and biodiversity.
Materials and Methods
Using published and unpublished records, we documented the worldwide range of B. chinensis. We obtained unpublished site records from specimens in the collections of the Smithsonian Institution National Museum of Natural History, the Museum of Comparative Zoology, the Mississippi Entomological Museum, the North Carolina State University Insect Collection, personal collections (in native and introduced ranges) and communications (particularly from people who suffered stings and contacted the lead author, B.G.). In addition, collection information on B. chinensis distribution was gathered through the use of the online database (Antweb version 7.10.4; www.antweb.org ) and through literature review.
Geographic coordinates for collection sites came from published references, specimen labels, maps, or geography websites (e.g., www.google.com/earth). However, many site names of older references and specimens, particularly in Asia, were obsolete or now spelled differently. If a site record listed a geographic region rather than a “point locale,” and no other record existed for this region, we used the coordinates of the largest town within the region or, in the case of small islands and natural areas, the center of the region. Records of B. chinensis found in newly imported goods or intercepted in transit by quarantine inspectors were excluded. For example, Forel (1900) reported B. chinensis (as P. solitaria) arriving in Hamburg, Germany, with plants shipped from Japan, and Wilson and Taylor (1967) noted that B. chinensis (as B. solitaria) was intercepted in quarantine in Honolulu, Hawaii.
REPRODUCTIVE PRODUCTION PERIOD
Nests were collected opportunistically from dead wood from 2007 to 2011 in North Carolina to estimate the sociometry of B. chinensis colonies. The presence of alate females and males, as well as brood (eggs, larvae, pupae), during nest collection was recorded to provide bionomic information on the reproductive phenology of B. chinensis colonies, with further data compiled from records found in literature and websites ( www.Antweb.org, BugGuide.net) for confirmation of the periods retrieved.
INFORMATIONAL WEBPAGE ON PACHYCONDYLA CHINENSIS
In 2009, the lead author created a webpage ( http://www4.ncsu.edu/~bsguenar/Pachycondyla%20chinensis%20page.html [inactive]) to provide information on Asian needle ant biology and how to distinguish them from other common ants. People who viewed this webpage sent the lead author emails detailing their experiences with this ant. Those email exchanges, obtained from Jul 2010 to Mar 2017, were archived and the results were presented anonymously. Everyone who contacted the lead author voluntarily initiated the communication after reading about the research conducted on B. chinensis. All records of B. chinensis were confirmed by seeing photographed specimens or actual samples sent by mail.
We compiled published and unpublished specimen records from 782 localities: 428 from the Old World and 354 from the New World. We report the earliest known B. chinensis records for 50 geographic areas in the Old World (36 countries, first-level administrative divisions, and island groups) and the New World (16 states in the US and Washington, D.C.).
In Asia, B. chinensis was reported from a large range, extending in latitude from North Korea (40.0000°N) to Java (6.6000°S), and in longitude from Nepal (87.1000°E) to the Ogasawara Islands (142.2000°E) (Fig. 1; Table 1), which most likely represents the extent of the B. chinensis species complex. The distribution of B. chinensis s.s. (based on the type specimen from Shanghai, most recent and accurate species records, as well as new material examined) was confirmed for Japan, Taiwan, and some of the eastern provinces of China, extending at least from Shanghai to Hong Kong (Fig. 1). However, its presence in Hong Kong was rare and confined to an area around the Kadoorie Farm and Botanic Garden (New Territories). While no specimens were directly examined from the Korean Peninsula, the presence of B. chinensis s.s. seems likely.
Other records, particularly those from Southeast Asia, need more careful examination, but it seems unlikely for this species to be found in the Philippines, Thailand, or in Java. Records from southern China and Vietnam will need to be verified because B. chinensis can be easily confused with B. obscurans which is slightly smaller in size. Nelder et al. (2006) wrote, “the Museum of Comparative Zoology (Harvard University, Cambridge, Massachusetts, USA) contains P. chinensis specimens collected from Guam, India, Indonesia, Myanmar, Nepal, Papua New Guinea, Philippines, Solomon Islands, Sri Lanka, and Thailand” (S. C., personal communication, Museum of Comparative Zoology). However, Stefan Cover told us that this was a miscommunication and that Nelder was referring to the range of all species in the B. chinensis species group. However, we were unable to find any specimens or records of B. chinensis from Burma (Myanmar), Guam, Solomon Islands, or Sri Lanka.
In its introduced range, B. chinensis was recorded along a continuous range on the East Coast of the US from Florida to Connecticut, west to Arkansas, and with 2 isolated records from Wisconsin and Washington State (Fig. 2; Table 2). Over the past 10 yr, numerous new records of this species have been added, and this species is now recorded in 16 US states plus Washington, D.C., spanning an area of > 965,000 km2. In the absence of standardized sampling efforts or detection programs, the conclusions that can be addressed are limited. However, the distribution pattern over time of B. chinensis shows a strong local establishment in Washington, D.C., Virginia, North Carolina, South Carolina, and Georgia, with populations recently confirmed in Alabama (J. A. M., personal observation) and Florida, and new populations detected in the past few years in Arkansas, Kentucky, Maryland, Mississippi, Ohio, Tennessee, and Wisconsin (Fig. 2; Table 2). In addition, some of the specimens from Kentucky and Mississippi were collected near the borders of Indiana and Louisiana (< 3 km). Collection records of this ant from western Arkansas were from only 1 county east of Oklahoma. Unfortunately, some of the northern records in coastal regions of Connecticut and New York were not available for examination and would thus require further confirmation. In the published literature, a collection record from New Jersey was mentioned (e.g., MacGown et al. 2013), but we could not find any records except for a quarantine specimen (Antweb CASENT0246023).
In the Old World, records of B. chinensis have been reported from shipments of plants in Hamburg, Germany, (Table 1) introduced from Japan; however, the identity of the specimens could not be confirmed. Records of B. chinensis recently have been reported from 4 localities bordering the Russian and Georgian coasts of the Black Sea (Table 1).
Earliest known records for Brachyponera chinensis Emery from the Old World. MCZ = Museum of Comparative Zoology. TARI = Taiwan Agricultural Research Institute. cu= collector unknown. *Indicates regions where species identification would require confirmation. +Indicates records which represent an introduced population.
Overall, most B. chinensis records were from sea level to 1,100 m asl, with a few records from higher elevations: 1,600 m asl; 1,620 m asl; 1,915 m asl; and a maximum elevation of 2,400 m asl in Nepal (M. C. Z., record collection) (Fig. 3).
From 49 nests collected, with collection records for every mo, the presence of alate females and males within B. chinensis nests was found to span continuously from early Apr to early Sept, while the production of brood was found continuously from early Apr to early Oct (Fig. 4). Because no collections were performed from mid-Feb to mid-Mar, we cannot exclude the possibility that reproductives and brood were produced during that time.
HEALTH RELATED PROBLEMS
From Jul 2010 to Jun 2017, 21 cases of health problems were reported to the authors after a sting by B. chinensis from Georgia, Kentucky, Mississippi, North Carolina, South Carolina, Tennessee, and Washington, D.C. Among those, at least 12 individuals reported having severe allergic reactions, including anaphylactic shock. The remaining individuals reported having symptoms such as moderate swelling, sweating, light-headedness, severe pain, stinging, and a burning sensation for a 2 h period. In at least 7 cases, multiple stings were reported, 4 of them associated with anaphylaxis. Several cases of stings happened during activities such as gardening or moving logs (which were used as nesting sites by B. chinensis). Other stinging events were by flying alate individuals falling into swimming pools or trapped underneath people's clothes. While we could not test this directly with the data available here, stinging events may be particularly frequent during summer swarming; 19 of 21 stinging cases were reported from Apr to Sep, with a peak in May to Jul (14 cases).
At the time of the first discovery of B. chinensis populations in the US in 1932, the species was already broadly distributed in the coastal regions of 3 states (Smith 1934). This distribution pattern suggests that the initial introduction of this ant species in the US might have occurred years earlier, perhaps in the early 1900s, or resulted from multiple independent introductions facilitated by human-mediated dispersal events from Asia to new regions of the New World. Indeed, there are several records of B. chinensis intercepted on imported plants (e.g., Prunus spp., Gentiana makoni) in the late 19th century (Forel 1900, Germany) and 20th century (Brown in 1965, New Jersey, Antweb record CASENT0246023). By 1950, B. chinensis was recorded from 6 US states, showing that the large scale spread of this species over large areas was not a recent phenomenon. However, the abundance of this species might have been very low and restricted to a few locations. For instance, Carter (1962), in 1 of the first lists of the ants of North Carolina, reported B. chinensis (as Euponera solitaria Smith) only from the eastern region of North Carolina known as the Coastal Plain (not including the sandhills region) based on 1938 records. Also in this report, Carter mentioned that he never collected this species himself, despite major sampling efforts in different habitats and parts of the state.
Recent records of B. chinensis from southern Russia (Krasnodar Krai) and Georgia (Abkhazia and Adjara regions) represent newly established populations within a new introduced region. Notably, all 4 records originate from coastal cities (Adler, Sochi, Sokhumi, and Batumi) along a 280-km front (Dubovikoff & Yusupov 2018; D. Dubovikoff, personal communication). These records were collected from botanical gardens (Adler and Bathumi), urban parks, as well as from protected natural habitats, and included both workers and queens (Yusupov collection; D. Dubovikoff, personal communication), supporting the presence of an established population in this region. These populations, along the coastal area of the Black Sea, resemble the initial 1932 records of B. chinensis in the US reported along the coastline of Virginia and North Carolina (Smith 1934). Important measures should thus be quickly developed to control the spread of B. chinensis and limit the risk of a population outbreak in this region. Bertelsmeier et al. (2013) developed models that identified the East Coast of the Black Sea (in Russia and the Republic of Georgia) and more western states of the southeastern US (e.g., Arkansas) as highly suitable areas on the basis of environmental conditions for B. chinensis population establishment. Similarly, the record from Bellingham (Whatcom County) in Washington State on the West coast of the USA would require further work to determine if it represents an independent introduction event directly from the native range of B. chinensis. This indicates that this species not only is increasing its invasive range within North America but also into new regions of the world. Important biosecurity measures for early detection should be considered in regions where similar suitable conditions are present, including countries along the Adriatic Sea, southern Brazil, Uruguay, northern Argentina, and New Zealand. In New Zealand, Brown (1958) reported the record of a single worker of B. chinensis from Waikino (Auckland Province) collected at an unknown date. In April 1959, Taylor (1961) collected 2 workers that he identified as B. luteipes from a timber yard in Penrose (Auckland Province). However, the author acknowledged the taxonomic confusion around southeastern Asian Brachyponera and hypothesized that the previous record from Brown was likely the same species. It is unclear whether the records collected in the late 1950s belonged to B. chinensis or to a similar introduced species. Because both localities were separated by about 100 km, this could suggest a potential widespread distribution at this period. Neither B. chinensis nor B. luteipes have been recorded from New Zealand for the past 60 yr. However, faunal ant surveys to reexamine those sites should be conducted for discovery of potential remnant populations, as similar circumstances existed for several decades after the initial detection of B. chinensis in the US mainland.
Earliest known records for Brachyponera chinensis Emery from the United States.
Unfortunately, few faunal ant studies were conducted in the US before the 1990s where B. chinensis may have been initially collected. This information would have allowed a clear understanding at which point abundance of this pest reached the high population levels observed today. Nonetheless, these results suggest a relatively prolonged lag time between introduction and ecological invasion. These 2 sets of occurrences have been characterized for many biological invaders (Crooks & Soulé 1999), including the red imported fire ant, Solenopsis invicta Buren (Hymenoptera: Formicidae). In this instance, the first established populations of S. invicta occurred in habitats corresponding to their native range but later spread to novel habitats within their introduced range (Fitzpatrick et al. 2007). However, the mechanisms that caused B. chinensis to suddenly reach high abundance are unclear, though diet diversification has been proposed recently as a potential result of ecological change compared with native ant populations (Suehiro et al. 2017). At a distributional range of over 965,000 km<sup>2</sup>, B. chinensis is now one the most widespread invasive ant species within the US; compared with S. invicta, the former has spread across an area of over 1.3 million km2 in the US (Williams et al. 2001). Currently, the temporal spread of several local B. chinensis populations is well documented illustrating the detrimental effects of this species on native ants in forested (Guénard & Dunn 2010; Warren et al. 2015; Suehiro et al. 2017) and urban habitats (Spicer-Rice & Silverman 2013). Basic understanding of this species' biology in its native and introduced ranges represent an important avenue that may lead to a management strategy to limit its spread locally (see example with Anoplolepis gracilipes, Hoffman 2015).
In addition, the well-publicized spread of the highly destructive S. invicta through the southeastern US has probably helped over-shadow the significance of B. chinensis, particularly regarding the public health issue. Every year, B. chinensis is responsible for several cases of strong allergic reactions, including anaphylaxis in the US (Leath et al. 2006). Within its native range in South Korea, an estimate of 2.1% of the human population present allergic reactions following B. chinensis stings (Cho et al. 2002). Allergic cases have been reported in Japan (Fukuzawa et al. 2002), and stinging has been reported from many parts of southern China (Srisong et al. 2016). Based on the results we obtained from voluntary communication from people stung by this ant coupled with those reported by Leath et al. (2006), it is suggested that the public health importance of B. chinensis might be greatly underestimated in the US. Moreover, the public health reputation of S. invicta and lack of public awareness regarding reactions to B. chinensis venom may have resulted in misdiagnosis of B. chinensis stings as fire ant stings. Interestingly, desensitizing treatments available for allergic reactions to fire ant stings appear ineffective in treating B. chinensis stings (Srisong et al. 2016). The combined association of the spread of B. chinensis with the sensitivity of human populations to venom from its sting represents a serious health concern that needs to be addressed by governmental institutions, particularly within the zone of invasion.
The authors would like to warmly thank the following persons for providing records information, specimens to examine, or assistance during field work: J. Boggs, J. Costa, R. Dunn, D. Dubovikoff, A. Lucky, K. Matsuura, S. Cover, K. Shimizu, M. Staab, K. Tsuji, E. Vargo, T. Yashiro, S. Yanoviak, and K. Yearly. The authors would also like to thank the sting victims of B. chinensis, who reported their experiences, pictures, and locality information used within this study. This research was supported in part by the National Institute of Food and Agriculture, Mississippi Agricultural and Forestry Experiment Station State Project MIS-012040, the USDA-ARS Areawide Management of Invasive Ants Project (Richard L. Brown, P.I.), and USDA-APHIS-PPQ Beetle Trapping Survey.
 Supplementary material in Florida Entomologist 101(4) (Dec 2018) is online at http://purl.fcla.edu/fcla/entomologist/browse