The Ants (Hymenoptera, Formicidae) of Sri Lanka: a taxonomic research summary and updated checklist

Abstract An updated checklist of the ants (Hymenoptera: Formicidae) of Sri Lanka is presented. These include representatives of eleven of the 17 known extant subfamilies with 341 valid ant species in 79 genera. Lioponera longitarsus Mayr, 1879 is reported as a new species country record for Sri Lanka. Notes about type localities, depositories, and relevant references to each species record are given. Accounts of the dubious and some undetermined species from Sri Lanka are also provided. 82 species (24%) are endemic whereas 18 species that are non-native to Sri Lanka are recorded. The list provides a synthesis of the regional taxonomical work carried out to date and will serve as a baseline for future studies on the ant fauna of this biodiversity hotspot.


Introduction
Sri Lanka is an island country located in the Indian Ocean, with latitudes ranging from 5°55'N to 9°51'N and longitudes from 79°41'E to 81°53'E. The country has a length of 432 km (268 miles) and a maximum width of 224 km (139 miles), with an area of 65,610 km 2 . Three distinct tropical climatic zones are prevalent in the region, the 'wet', 'dry', and 'intermediate zones' based on seasonal precipitation distribution. These regions receive more than 2,500 mm; less than 1,750 mm, and between 1,750 to 2,500 mm of rain respectively with average annual temperature ranging from 28 °C to 31 °C (Karunaweera et al. 2014).
The island is part of the same shallow continental shelf as India, and is only separated by an inlet of the Bay of Bengal known as the Palk Strait (Pathirana 1980;Somasekaram 1997). This 40 to 85 mile-wide and approximately 85 miles-long strait separates southeastern India and northern Sri Lanka (Vaz 2000). Historically, with high faunal affinities observed across the Palk Strait, the concept of Ceylonese or Lankan biogeographic region was conceived (Wallace 1876;Chandran 1997). More recently, both areas were considered together as the Western Ghats -Sri Lanka biodiversity hotspot, representing a single seasonal wet region in the whole of South Asia (Myers 1988;Mittermeier et al. 2004). This region on the basis of three factors: high numbers of endemics and endemic species/area ratios for both plants and vertebrates, and habitat loss is considered as one of the main hotspots of the world (Myers et al. 2000). Despite several extended periods of land connection during the past 500,000 years, Sri Lanka has maintained a fauna that is largely distinct from that of the Indian mainland (Bossuyt et al. 2004). Unfortunately, this pattern has been tested for only a limited number of taxa in plants, vertebrates, or invertebrates, due to the limitation of data available. The proportion of endemic species in plants is ~ 25%, vertebrates ~ 30%, and invertebrates ~ 43% in the groups studied in depth (Bossuyt et al. 2004; Gunawardene et al. 2007; Gunatilleke et al. 2008). Such variation in endemism and its understanding at finest possible scales will help to develop conservation management programs for the entire region (Gunawardene et al. 2007;Dad et al. 2019). Sri Lanka is known for its remarkable biodiversity and considered to be one of the richest countries in the Asian region in terms of species concentration with regard to mammals, reptiles, amphibians, fish, and flowering plants (NARESA 1991). The highest species diversity is recorded among the flowering plants (3771 species), followed in decreasing order by the fungi (~ 2260), bryophytes (788), freshwater algae (~ 560) and ferns (314) (Gunatilleke et al. 2008). Among animals, the diversity of vertebrates is well known in comparison to that of invertebrates, where only a few groups have been studied in depth (Gunatilleke et al. 2008). With rapidly decreasing forest cover (Mattsson et al. 2012) more and more species are increasingly threatened; with to this date 27% of birds, 66% of amphibians, 56% of mammals, 49% of freshwater fish, 59% of reptiles, and 44% of flowering plants classified as threatened under the IUCN Red List categories (MOE 2012;CBD 2020). The limited number of trained taxonomists, lack of initiative to explore the biodiversity and the loss of primary forest cover are currently the biggest drawbacks in the conservation of biodiversity in Sri Lanka (Bawa et al. 2007). In particular, knowledge on the entomofauna of Sri Lanka is particularly limited, with comprehensive species checklists only available for a handful of taxonomic groups: such as bees, butterflies and Odonata (Karunaratne et al. 2012;van der Poorten 2012;van der Poorten and Conniff 2012).
In recent decades, Sri Lankan authors have contributed to the taxonomic and ecological study of ants including: Dias ( , 2006a, Gunawardene et al. (2008Gunawardene et al. ( , 2010Gunawardene et al. ( , 2012, , Dias and Rajapaksa (2016), Dias et al. ( , 2018 and Yamane and Dias (2016). The information available on ants in Sri Lanka is, however, mainly restricted to a few districts and largely confined to the 'wet zone'. It is thus highly likely that other climatic zones, which have received less atten-tion, may contribute significantly to the overall regional ant fauna, with possibilities of many new discoveries once properly surveyed.
It should also be noted that several exotic ant species with rampant effect on native mesofauna but mostly undocumented and unappreciated are established in Sri Lanka . To date, there is no updated and annotated checklist of Sri Lankan ants available and therefore, the objective of the present study is to provide a comprehensive checklist of Sri Lankan ant species, and to highlight gaps where additional faunistic surveys and research are needed to fully understand the diversity of this group in the region.

Data sources
The checklist is primarily based on available literature and few museum records. Most of the names of described species presented are in accordance with the most recent classification following Bolton (2020). Important references to species records are provided.
Species records are presented in function of their mention of examined material within the published publication (primary literature records; e.g., specimen of Technomyrmex albipes examined by Bolton and published in Bolton 2007), repetition of known records from other publication (secondary literature records; only the record of a previous publication is being referred to within addition of new material examined), or on the basis of specimens examined here (material examined), inclusive of specimens available on AntWeb (AntWeb records).
The exploration of the Sri Lankan ant fauna and descriptions of species has, however, been relatively limited during the past few decades. For instance, more than a century ago, by 1920, 66.8% of the diversity currently known had already been recorded (Fig. 1). This number increased slowly for 80 years with the addition of 69 newly recorded species (19.9% of the total fauna); and slightly faster in the most recent 20 years with an addition of 46 species (12.7%) mainly through the work of local scientists (Fig. 1A, B). The taxonomic work still required on Sri Lankan ants is likely to be important due to the high number of species and subspecies recorded more than a century ago, at a time when species descriptions were sometimes incomplete or species boundaries poorly defined, later leading to potential misidentifications (see Table 4). As a result, work on regional exploration and the use of new sampling are both needed, as shown by the paucity of certain diverse genera (e.g., Colobopsis, Myrmecina, Stigmatomma, Strumigenys), while taxonomic revisions of specimens for the region based on newly collected material to confirm past identifications or to help in the descriptions of new species should help in providing a more complete overview of the diversity of the Sri Lankan myrmecofauna. Of the three traditionally distinct climatic zones in Sri Lanka: 'dry' (~ 40669 km 2 ), 'intermediate' (~ 9670 km 2 ) and the 'wet zone' (~ 15267 km 2 ), most of the ant surveys and species occurrence have been reported from the 'wet zone' areas, which include some of the well protected and intact forest regions of the country ( Fig. 2; Suppl. material 2). The 'wet zone' supports the greatest diversity (231 species), followed by the 'dry' (113 species), and the 'intermediate zone ' (71 species). It should be noted that most of the northern and eastern regions of the country, located in the dry zone, have received limited sampling coverage, and that future surveys should focus on these regions. Sixty-four species recorded from Sri Lanka could not be assigned to any zone as they lack precise location information within the country underlying, here again, the paucity of surveys since the time of their original recording in Sri Lanka and the need for future and further research in the various regions of the country.

Endemic species
Sri Lanka is known to have diverse vegetation types and a distinctive fauna characterized by a high degree of endemicity (Gunawardene et al. 2007;Gunatilleke et al. 2008). Of the 341 ant species/subspecies present in Sri Lanka, only 82 species (24%) are considered as endemic (Table 2). Thanks to recent global compilation and regional work on ant distribution, the number of species identified as endemic species to Sri Lanka has greatly increased in comparison of previous work in which only 33 species (17% of 194 species) had been identified . It should be noted, nonetheless, that the current level of endemism retrieved in ants is relatively low in comparison of flowering plants (28%), odonates (48%), reptiles (59%), land snails (83%), and amphibians (85%) (Gunawardene et al. 2007;Gunatilleke et al. 2008;MOE 2012). Potentially, a similar pattern of high species richness and endemism might be likely for ants, but further comprehensive surveys in addition to thorough taxonomic work are first needed to Figure 1. A Rate of species recording in Sri Lanka per decade from 1886 to 2020, and B number of species recorded in Sri Lanka per period in function of the origins of the authors contributing to these discoveries. The last two decades are shown separately as they present a change in species discovery (two species are not included as time of discovery is unknown).
fully depict this pattern. However, the current legal framework for biodiversity conservation allowing the export of specimens outside Sri Lanka is very restrictive and might represent a serious limitation to the completion of the taxonomic work that could be undertaken within the country. Moreover, with ongoing landscape modification in the region, many species have been driven to critical status (Somaweera et al. 2015;Perera et al. 2017;Karawita et al. 2018), therefore, urgent, large-scale, and sustained efforts to monitor, characterize, and conserve the ant fauna of Sri Lanka is critical.

Non-native species
The current list of introduced and established species in Sri Lanka includes 18 species largely dominated by the Myrmicinae (13) and completed by the Ponerinae (3) and Formicinae (2) subfamilies ( Table 3). The ecological impacts of non-native and invasive ant species with rampant effect on native mesofauna have not been studied in Sri Lanka, but the well-known effects of some of these species in other regions of the world (Wittman 2014), including within Asia (Wong et al. 2020), may hint towards similar outcomes. Therefore, more efforts should be directed to evaluate the distribution, ecology, and various impacts of non-native species. Potential threat of non-native species on native species, particularly endemics should be evaluated to safeguard the native ant fauna. Finally, the limited efforts in the study of Sri Lankan ants suggests that more non-native species could be discovered in future studies, or through biogeographic studies aiming at determining more precisely the native from the introduced ranges of species widespread within Asia and beyond.

Misidentifications and dubious/erroneous records
To correct the errors cited in earlier literature so as to reduce the taxonomic confusion by eliminating misinformation associated with Sri Lankan ants, 58 ant taxa previously reported from the country are here marked as dubious based on either erroneous data in terms of misidentifications, misspellings, erroneous locality, or potential occurrence. A brief explanation is provided about their dubious status (Table 4). Table 3. List of non-native ants in Sri Lanka. Species with an asterisk * are considered as invasive in other regions of the world.