First record of the cryptic invader Pyrgophorus platyrachis Thompson , 1968 ( Gastropoda : Truncatelloidea : Cochliopidae ) outside the Americas

Anthropogenic removal of natural dispersal barriers and modification of natural habitats have contributed to the spread of non-native species. Potential invaders that are cryptic in appearance and/or behaviour are particularly troublesome as this confounds efforts to detect or manage incipient invasions. Here we report one such invader, the Florida serrate crownsnail, Pyrgophorus platyrachis Thompson, 1968, in Singapore reservoirs—only the first record outside the Americas. We identified P. platyrachis using morphological and molecular techniques. The Singapore COI mitochondrial sequences were 99–100% matched to P. platyrachis from Florida, USA, while the 16S sequences are the first published for this species, and were most closely matched to another confamilial species.


Introduction
The removal of natural dispersal barriers and the creation of empty or novel niches in modified natural habitats during the Anthropocene (McKinney 2006;Hulme 2009) has facilitated the propagation and establishment of non-native species in a range of habitats resulting in both direct and indirect impacts (Clavero and García-Berthou 2005;Nghiem et al. 2013).Although biological invasions are nearly ubiquitous, especially in disturbed habitats, not all invaders are equal, with species traits such as high fecundity and generalist feeding habits increasing the likelihood of establishment in adventive habitats (Moyle and Marchetti 2006).These model invaders become an even greater concern if they are difficult to detect as this negates our ability to pro-actively or rapidly address an incipient invasion before eradication becomes a practical impossibility (Bertolino and Lurz 2013).
A group with all the traits of an ideal invader is the New World freshwater snail genus Pyrgophorus Ancey, 1888 (Truncatelloidea: Cochliopidae), which is distributed in Florida, the Caribbean islands and Texas southwards to Mexico until Venezuela (Thompson 1968;Hershler and Thompson 1992;Pointier 2015).Pyrgophorus snails are known to occur in high densities (up to >15,000 individuals m -2 [Dillon et al. 2006]), and capable of live-bearing relatively large numbers of offspring (i.e., 50 per batch) (Thompson 1968).These snails have spread beyond their native ranges, but none have been definitively identified to species owing to the taxonomic complexity of the group (Englund 2002;Mienis et al. 2011;Nasarat et al. 2014).The inability to conclusively identify members of the genus contributes to the cryptic nature of their invasions, particularly in habitats where superficially similar native snails occur.Adding to the confusion, very little is known about the ecology of these cryptic invaders, with physicochemical environmental parameters recorded for a single Pyrgophorus species in Venezuela (Nava and Severeyn 2011) representing all we know about the environmental tolerance of this group.This knowledge gap can be hugely detrimental in the assessment of the potential spread and impacts of members of this genus.The sheer densities of their populations may be indicative of their potential invasive impacts.An analogous invasion case study is present in the similarly-sized New Zealand mud snail Potamopyrgus antipodarum (Gray, 1843), a globally invasive species that reaches similar densities and is known to impact native aquatic communities via competitive exclusion, predation, and habitat modification (Alonso and Castro-Díez 2012).
In a perfect illustration of the high levels of global connectivity in the "New Pangaea" (Baiser et al. 2012), we report the first record of the Florida serrate crownsnail, Pyrgophorus platyrachis Thompson, 1968, in Singapore, an equatorial island in Southeast Asia.Singapore's freshwater habitats comprise mainly natural forest streams, and urban man-made canals and reservoirs (Yeo et al. 2010).Freshwater snails in Singapore are found almost exclusively in urban, disturbed habitats-with man-made reservoirs having the highest species richness (Clements et al. 2006;Tan et al. 2012).Although some species are native to the surrounding Southeast Asian region, most of the freshwater snails in Singapore are considered non-native or are at least cryptogenic owing to their exclusive association with human-disturbed habitats (Tan et al. 2012).Pyrgophorus platyrachis is native to Florida (Thompson 1968), with a doubtful occurrence recorded in Venezuela (see Pointier 2015), making the established population in Singapore the first to be recorded outside the Americas.We first discovered P. platyrachis in Singapore on 31 October 2014 in inland reservoirs and in this study we include mitochondrial sequence analysis to confirm the species' identity.

Methods
We recorded P. platyrachis from three reservoirs in Singapore-Bedok, Pandan, and Punggol Reservoirs (Figure 1, supplementary material Table S1).All three are artificial habitats located in urbanised areas: Pandan Reservoir was created over mangrove swamps at the mouth of the Pandan River; Bedok Reservoir is a former sand quarry; and Punggol Reservoir was formed by damming the previously estuarine Punggol River (Yeo et al. 2010).To characterize their habitat and environmental tolerance, physicochemical variables-water temperature (°C), pH, dissolved oxygen (mgL -1 ), conductivity (mScm -1 ), and total dissolved solids (gL -1 )-were measured using a YSI Professional Plus handheld multiparameter meter (YSI Inc.).
We collected P. platyrachis specimens using customised, cage-type samplers (diameter 20.0 cm; height 10.0 cm, 1.3 cm 2 mesh size) as artificial invertebrate colonisers (Loke et al. 2010) and a 15.2 cm 3 Ekman grab (Forestry Suppliers, Inc.).Six colonisers were deployed per reservoir, in areas with soft, muddy substrata at depths of 0.5-2.0m for two to three weeks.All samples were collected between October 2014  S1).
and June 2015, and deposited in the Zoological Reference Collection (ZRC) of the Lee Kong Chian Natural History Museum, National University of Singapore (Catalogue numbers ZRC.MOL.005854-005861).
Over 1000 snails per coloniser from Pandan and Punggol Reservoirs, and two individuals from Bedok Reservoir were examined.In order to facilitate further identification, we separated the body of selected snails from the shell by placing the specimens inside a microwave oven (800W) for 10-15 seconds (Galindo et al. 2014).The shell and body (including penial morphology) were examined using a stereomicroscope.Identification of species was based on Thompson (1968;2004).

Morphological identification
The species is diagnosed by the following morphological characters: shell ovate-conic, spire elongated, suture deeply impressed, with 4-6 whorls, peristome slightly detached and complete in mature individuals, dextral; operculum elongate-ellipsoidal, paucispiral, thin and flexible (Figure 2A); penis with 4 papillae on the right margin, tip tapered and pigmented, two papillae on the left margin, immediately below the tip, smaller papilla on the proximal left margin (Figure 2C) (sensu Thompson 1968;2004).Living snails were semi-transparent and yellow-speckled (Figure 2B).Three distinct morphs were recorded: i) relatively smooth-shelled, ii) shells with raised spiral threads, and iii) shells with spines on the periphery of the threads (Figure 2A).Shell height 2.5-4.6 mm; shell width 1.5-2.5 mm.

Molecular analysis
In addition to morphological assessment of the specimens, we also successfully sequenced both COI and 16S genes from two individuals, one individual each from Pandan and Punggol Reservoirs.The COI sequences (KT326190 and KT372180) matched Pyrgophorus platyrachis (99-100% identity) AF129327 and AF367632 (Hershler et al. 1999;Wilke et al. 2001), while both 16S sequences (KT326188 and KT326189) were most closely matched to the confamilial Spurwinkia salsa (Pilsbry, 1905) (97% identity) EU573991 (Ponder et al. 2008).Based on objective clustering of five COI sequences (two from this study, the two top hits from GENBANK, and S. salsa AF367633 [Ponder et al. 2008]), two genetic clusters were formed.The Singapore individuals and the published P. platyrachis sequences from GENBANK formed one cluster, while S. salsa clustered differently, separated by a 6.6% divergence.For 16S, both sequences from this study clustered together, while the published sequence of S. salsa recorded a divergence value of 3%, similar to the BLASTN results.No other published P. platyrachis 16S sequences were available for comparison.

Species identification
Identifying members of Pyrgophorus is problematic, with populations introduced to the Middle East and Hawaii yet to be definitively identified to species level (Englund 2002;Mienis et al. 2011;Nasarat et al. 2014).The taxonomic confusion is exemplified by the identification of P. parvulus in Venezuela as a junior synonym of P. platyrachis by Nava et al. (2011) though others consider P. parvulus a valid species (Pointier 2015).Here we used a combination of morphological and molecular evidence to verify the identity of the present aquatic snail species belonging to a complex and taxonomically unresolved genus (Thompson 1968;Hershler and Thompson 1992) to confirm the first record of P. platyrachis outside the Americas.The three distinct shell morphs are typical of the genus (Hershler and Thompson 1992), and were similar to the morphs recorded from Florida (Thompson 2004).The top hits for the COI gene also supported the identity of the species as P. platyrachis.The close match of the 16S gene to the confamilial S. salsa is not unexpected given that interspecific variability can differ very widely, and nuclear markers would be required to better delimit the cochliopids (Meier 2008).
In-situ identification of newly introduced species can be difficult in the presence of morphologicallysimilar native or established species.Pyrgophorus platyrachis may have first been collected as early as 2013 (JHL pers.obs.), but were likely mistaken for juveniles of the native Thiaridae (Figure 3), which are common in Singapore's fresh waters (Tan et al. 2012).Thiarids here, however, can be easily distinguished from Pyrgophorus spp.by the presence of prominent shell sculpture, e.g., axial ribs and/or spiral grooves (versus absence of obvious shell sculpture in Pyrgophorus) (Figure 3), and incomplete peristome interrupted at the parietal area (versus the slightly detached and complete peristome seen in mature Pyrgophorus individuals) (Figure 2A).Cochliopidae are also superficially similar in size and shape to the globally-invasive New Zealand mud snail, Potamopyrgus antipodarum (not recorded in Singapore to date), and recently introduced cochliopids elsewhere have been mistakenly identified as the mud snail (Hershler et al. 2015).The quickest way to distinguish these species is by comparing morphs with spinesspines of Potamopyrgus are not calcified and are part of the periostracum, while spines of Pyrgophorus are protrusions of the shell itself (Winterbourn 1970;Mienis et al. 2011).

Habitat characteristics
We encountered the highest abundances of P. platyrachis at sites with muddy bottoms.This is consistent with the snail's typical habitat in its native range in Florida, where they are most commonly found on floating plants and in fresh waters with soft, muddy sediment, although some populations are known to occur in mangroves (Thompson 1968).While there appears to be no available environmental variables recorded from its native habitats in Florida for comparison, ecological studies have been conducted on a congener, P. parvulus, in Venezuela (Nava and Severyn 2011 [as P. platyrachis]; Pointier 2015), and the physicochemical conditions of all the P. platyrachis collection sites in Singapore are within the pH and dissolved oxygen ranges of the sites in Venezuela (Nava and Severyn 2011).Our findings do not reflect the plasticity in habitat preference or tolerance of environmental variables sometimes reported in biological invaders.

Possible routes of introduction
We did not find evidence to shed more light on the introduction pathways of P. platyrachis into Singapore.Birds are known vectors of invasive snails (Hershler et al. 2015), but P. platyrachis has not been recorded anywhere else along the East Asia-Australasian Flyway, the path taken by migratory birds passing through Singapore (Wang and Hails 2007).Therefore, we postulate that the snails may instead have been accidentally introduced to Singapore via the ornamental trade, a key introduction pathway for aquatic plants and animals in Singapore (Yeo and Chia 2010).In both native and introduced habitats, Pyrgophorus species have been reported to be associated with floating macrophytes like water hyacinth Eichhornia crassipes (Mart.)Solms, and water lettuce Pistia stratiotes L. (Thompson 1968;Mienis et al. 2011;Nava et al. 2011)-both of which have been introduced into Singapore (Yeo et al. 2010).It is possible that the snails were inadvertently brought in with ornamental aquatic plants and introduced indirectly via indiscriminate disposal of aquarium water into the waterways (Duggan 2010), or planting of ornamental plants for landscaping purposes in the vicinity of water bodies.

Implications and conclusions
Owing to its propensity to be undetected because of its small size and unremarkable appearance, the actual introduced range of Pyrgophorus platyrachis outside its native distribution may be wider than current literature suggests.Moreover, even when noticed, the snails may be mistaken for another morphologically similar invasive species that is more widespread and well-known, or even juveniles of native species.Our study highlights the importance of rapid identification of potential invaders using a combination of morphological and molecular assessments as this, along with the ecology of the organism in adventive habitats, can be crucial in managing and preventing the further spread of biological invaders (Pyšek et al. 2013).

Figure 1 .
Figure 1.Known distribution of Pyrgophorus platyrachis in Singapore (see details in supplementary material TableS1).