A review of the scopelocheirid amphipods (Crustacea, Amphipoda, Lysianassoidea), with the description of new taxa from Australian waters

Indexing Information Biological Abstracts® (Thompson ISI) BIOSIS Previews® (Thompson ISI) Camgridge Scientific Abstracts (CSA/CIG) Web of Science® (Thompson ISI) Zoological RecordTM (Thompson ISI) museum für naturkunde A review of the scopelocheirid amphipods (Crustacea, Amphipoda, Lysianassoidea), with the description of new taxa from Australian waters Niamh M. Kilgallen1, James K. Lowry1 1 Australian Museum Research Institute, 6 College Street, Sydney, NSW 2010, Australia http://zoobank.org/CAFFC884-904F-40C2-AACF-12BE3A2F3ECC Corresponding author: Niamh M. Kilgallen (niamh.kilgallen@austmus.gov.au)


Language and style
• The language of publication is English. There is no general limitation of the length of manuscripts, but please contact the editor before submitting papers exceeding 30 printed pages (approximately 60 manuscript pages including figures). • Manuscripts should be written in a clear, straightforward style and must not have been published or submitted elsewhere. • The text should be 12 pt, double-spaced, one-sided, left justified and with a margin of at least 3 cm. • Use a standard typeface, e.g. Times New Roman as little formatted as possible (without tabulators, several blank spaces, etc.). Avoid footnotes. • Divide the text into sections using headlines and subheadlines. Do not number the headlines. Inline headers should be set in italics and followed by a full stop. • The names of genera and species must be in italics.
• Taxonomic descriptions must comply with the rules of the 4th edition of the ICZN (see http://www.iczn.org/). • Enter the page number on every page. • Submit figures with a minimum resolution of 300 dpi. • The preferred file formats are PSD (Photoshop) and TIFF for colour and grayscale illustrations, and EPS for vector graphics. • JPG files are only accepted in high resolution.

General manuscript structure
If appropriate, the manuscript should be structured using headlines and sub-headlines, but without numbering, according to the following sections: -Title page -Abstract -Introduction -Materials and Methods -Results -Discussion -Acknowledgements -References -Tables with captions - Figure captions Introduction Scopelocheiridae Lowry & Stoddart, 1997 is a small family of scavenging lysianassoid amphipods which contains two subfamilies, Scopelocheirinae subfam. n. and Paracallisominae subfam. n. The scopelocheirines contain three genera and eight species living in temperate and boreal, mainly shallow, waters of the Mediterranean Sea, the North and South Atlantic, Japan and Australia. They are scavengers feeding on carrion on the sea bottom, with some species reported feeding in the dead tests of spatangoid urchins (Chevreux 1911, Lowry andStoddart 1989). Paracallisomines are a larger group (seven genera and 15 species) widespread in the deep-  Gurjanova, 1962. Discussion. Stroobants (1976) considered the possibility that the specimen described by Gurjanova (1962) is a juvenile of Paracallisoma alberti. We consider it to be a valid species and placed in the subfamily Scopelocheirinae based on the columnar molar. We examined the syntype in the Australian Museum, but were unable to determine whether the molar has a triturating surface.
Ecology. Scopelocheirus species are frequently taken in baited traps. They are also reported in the literature as an associate of echinoids.

Ecology.
A scavenger that has been collected in baited traps (Chevreux 1935) and from dead fish on fishermen's lines (Sars 1890).
Type material. Probably lost.

Distribution. Widely recorded from the North Atlantic
Ocean and Mediterranean Sea. Records from Japan are tentatively referred to Aroui onagawae (Sekiguchi & Yamaguchi, 1983). North Atlantic Ocean. Denmark: near Horns Revs Lighthouse (Stephensen 1923b); the eastern Skagerrak (Stephensen 1923b); north of Skagen (Enequist 1949). France: Fosse de Capbretton (Norman 1900); Bay of Biscay (Chevreux 1903); off Roscoff (Dauvin 1988, Dauvin et al. 1994; Cap-Ferrat Canyon (Dauvin and Sorbe 1996). Guinea-Bissau: (Mateus and Mateus 1986). Ireland: Off the Skelligs, Co. Kerry; Ballycotton, Co. Cork  (Kaartvedt 1989); Norwegian Shelf area (Buhl-Mortensen 1996); the Skagerrak (Miskov-Nodland et al. 1999). Portugal: off Aveiro (Andres et al. 1992, Cunha et al. 1997); south of Olhão (Castro et al. 2005). Sweden: Bohuslän (Bruzelius 1859) (as Anonyx Kröyeri); Gullmar Fjord, Bohuslän (Enequist 1949, Oldevig 1959, Buhl-Jensen and Fosså 1991; west of Hållo (Enequist 1949); Löken, Gåsö Ränn; west of Nidingen (Oldevig 1959 (Chevreux 1935); Strangford Lough, Co. Down; off Donaghadee, Co. Down (Williams 1954); off Blyth, Northumberland (Bossanyi 1957); Clyde Area (Sanderson 1973, Moore 1984, Bergmann et al. 2002; near Assynt (Sanderson 1973); Anglesey (Ramsay et al. 1997). Discussion. Japanese records of S. hopei by Nagata (1965) and Sekiguchi and Yamaguchi (1983) are here considered to be inaccurate on the basis of their distribution, and some morphological inconsistencies with the European form such as the shape of the urosomite 1 (dorsally rounded in the European S. hopei, dorsally truncated in the Japanese specimens) and the length of the uropod 3 inner ramus (slightly shorter than and reaching at least to article 2 of outer ramus European specimens, much shorter than outer ramus in the Japanese specimens). It is possible that these records may actually represent Aroui onagawae. Unfortunately, the only illustration of Japanese specimens (by Sekiguchi and Yamaguchi (1983)) does not show the setae on the outer plate of maxilla 2 and as such precludes a generic placement in either Aroui or Scopelocheirus. However, all of the other illustrated characters correspond to the description and illustration of A. onagawae by Takekawa and Ishimaru (2000).  Discussion. Anisocallisoma can be distinguished from all other paracallisomines by the reduction in the number of setae of the maxilla 1 inner plate. It is most similar to Eucallisoma Barnard, 1961, and Tayabasa gen. n. They share the following characters: gnathopod 1 basis swollen, glandular; dactylus reduced, simple. It is also very similar to the new genus Austrocallisoma, but it can be distinguished from all of these taxa in lacking the distal tuft of setae on the accessory flagellum, and in having a much more weakly-developed posteroventral lobe on the pereopod 4 coxa, as well as the reduced setae on the maxilla 1.
Etymology. The name is a combination of the prefix Austro-from the latin australis, meaning southern and referring to the southern hemisphere distribution of the type species, and the suffix -callisoma (gender neuter) referring to its placement within the Paracallisominae.
Discussion. Austrocallisoma gen. n. is a difficult taxon that has much in common with the monotypic genera Anisocallisoma, Eucallisoma and Tayabasa gen. n. Having four monotypic genera that are highly derived yet clearly closely related is not ideal. However, to maintain consistency of diagnostic characters at a generic level we feel justified in establishing this new genus. Austrocallisoma can be separated from both Eucallisoma and Anisocallisoma by the strongly developed and subacute posteroventral lobe on the pereopod 4 coxa (well-developed and subquadrate in Eucallisoma, very weakly-developed and subacute in Anisocallisoma). It can be further distinguished from Anisocallisoma in having plumose setae lining the inner margin of the inner  Diagnosis. Mandible lacinia mobilis a stemmed distally expanded, irregularly cusped blade. Maxilla 1 palp 1-articulate. Maxilliped palp article 4 absent. Gnathopod 1 coxa reduced, slightly shorter than coxa 2; basis swollen, without glandular material.
Etymology. Named in honour of Jerry Barnard, in recognition of his enormous contribution to amphipod taxonomy.  Discussion. The tip of the outer ramus on uropod 3 on both sides is damaged in both specimens available for study. Judging from where the damage occurs we suspect that the ramus is 2-articlulate, however this is uncertain.
Discussion. Dahl (1959) Birstein and Vinogradov (1958) published an account of the amphipods of the north-western Pacific, including a new species, Scopelocheirus schellenbergi, also with aff. Paracallisoma spec. Schellenberg 1955 in its synonymy. Dahl (1959) consequently included a footnote in his account, stating that Schellenberg's specimen should be referred to Scopelocheirus schellenbergi, which in turn should be recombined as Bathycallisoma schellenbergi. He considered his Kermadec specimen to be a separate species from B. schellenbergi based on the shape of the first gnathopod and "some other minor characteristics". We cannot observe these differences and so prefer to retain B. pacifica as a junior subjective synonym of B. schellenbergi, thereby agreeing with most subsequent authors.
Included species. Eucallisoma includes one species: E. glandulosa J.L. Barnard, 1961. Discussion. The removal of E. barnardi Lowry & Stoddart, 1993 to Tayabasa gen. n. leaves Eucallisoma as a monotypic taxon. Future deep-sea samples will hopefully uncover associated taxa that will provide a clearer picture of the relationships between these animals.
Distribution. Gabon: west of Nyanga Province.

Type species. Paracallisoma alberti
Discussion. In addition to the two new species described herein, Horton et al. (2013) record an additional five undescribed species of Paracallisoma from the North Atlantic Ridge, and Duffy et al. (2012) record two undescribed species from submarine canyons of the Iberian Peninsula. These records indicate that there is still a large knowledge gap in the diversity of deep-sea scopelocheirids. Scopelocheirus abyssi Oldevig, 1959: 16, figs 1-3. -Barnard and Karaman 1991: 528. -Vinogradov et al. 1996: 8. -Brandt 1997: 1540
Discussion. The distribution of Paracallisoma alberti given here (Fig. 22) is much more limited than what has been reported in the literature. This follows Thurston (1990), who concluded that P. alberti is restricted to the north-east Atlantic, as material recorded in the literature as P. alberti from the Pacific Ocean was confirmed as or presumed to belong to P. coecum (Holmes, 1908). The identity of material from the Indian Ocean and Arabian Sea (Birstein andVinogradov 1964, Treude et al. 2002) is unknown, but it is unlikely to be P. alberti. Material from the Southern Ocean appears to be closely related to P. platepistomum Andres, 1977(Thurston 1990.   Discussion. Schellenberg (1926) first considered Paracallisoma coecum to be a junior subjective synonym of P. alberti, a move that was accepted by many subsequent authors. However, Barnard (1964), and many more recent works (e.g. Thurston 1990, Karaman 1991, Thurston et al. 2001) have again treated P. coecum as a valid species, a decision with which we agree. Paracallisoma coecum can be differentiated from P. alberti by following characters: gnathopod 1 coxa much longer than wide, margins slightly tapering distally (coxa slightly longer than wide, margins subparallel in P. alberti); gnathopod 1 propodus margins tapering distally (gnathopod 1 propodus margins subparallel in P. alberti); gnathopod 2 propodus subovate, palm transverse to slightly acute, dactylus fitting palm (gnathopod 2 propodus diverging distally, palm acute, dactylus shorter than palm in P. alberti). Due to its taxonomic history, many records of P. coecum have erroneously been attributed to P. alberti. Pacific Ocean material reported as P. alberti has now been confirmed as or is presumed to be P. coecum (Thurston 1990). According to Thurston (1990), the record of Shoemaker (1945) of P. coecum (as Scopelocheirus coecus) from Bermuda is referable to P. platepistomum Andres, 1977. Indian Ocean material recorded as P. alberti is unconfirmed.
Etymology. Named for Woolgoolga, a town west of the type locality on the coast of New South Wales; used as a noun in apposition.

Distribution.
Australia: east of Mooloolaba, Queensland, to south of Tasmania.

Ecology. A scavenger taken in baited traps.
Discussion. Paracallisoma woolgoolga sp. n. is morphologically very close to P. spinipoda. It can be distinguished from that species by the gnathopod 2 palm (slightly concave in P. spinipoda, straight in P. woollgoolga); the shape of the pereopod 5 basis (evenly rounded in P. spinipoda, with a slight excavation along the posteroventral margin in P. woolgoolga); and the shape of the epimeron 2 posteroventral corner (producing a small spine in P. spinipoda, subquadrate in P. woolgoolga). In addition the pereopod 6 basis is much less distinctly excavate posteriorly in P. woolgoolga compared with that of P. spinipoda. imens, AM P.48127,   Diagnosis. Gnathopod 1 coxa margins subparallel. Gnathopod 2 propodus palm transverse, with slightly concave, minutely serrate margin; dactylus reaching corner of palm. Pereopod 5 basis much longer than broad; basis slightly to moderately expanded posteriorly, posterior margin straight. Epimeron 3 posteroventral corner produced into a weak spine. Telson deeply cleft. Description. Based on holotype, male, 12.0 mm, AM P.69091.
Etymology. The species is named for Bert Ziviani, skipper of the RV Sunbird.
Distribution. Australia: east of Flynn Reef, Queensland, to north-east of Coffs Harbour, New South Wales.

Ecology. A scavenger, taken in baited traps.
Discussion. With its strongly developed pereopod 4 posteroventral lobe and relatively narrow and posterodistally lobate pereopod 5 basis, this species is most similar to P. alberti, P. platepistomum, and P. coecum. It can be differentiated from the latter two species by the shape of the gnathopod 1 coxa, which is short with subparallel margins (longer than broad and tapering distally in P. platepistomum and P. coecum) and the shape of the pereopod 7 basis (more distinctly excavate posteriorly in P. zivianii sp. n.). It differs from P. alberti in the shape of the gnathopod 2 palm, which is transverse, and the dactylus, which fits the palm (palm acute, dactylus distinctly shorter than the palm in P. alberti). Pereopod 4 coxa with weakly-developed, subacutely produced posteroventral lobe.

Scopelocheiropsis
Discussion. Scopelocheiropsis has some variable characters, most importantly the absence of a molar in S. sublitoralis (present in the both S. abyssalis and S. armata), and the blunt, reduced maxilliped palp article 4 in S. abyssalis (well-developed in the other two species). Nevertheless, the distinctively compressed carpus of pereopods 3 and 4, as well as the stemmed and distally expanded lacinia mobilis are strong diagnostic characters which separate these taxa from other groups. Discussion. Scopelocheiropsis abyssalis was originally described and illustrated as lacking a mandibular molar. Hendrycks and Conlan (2003) described new material and indicated the presence of a small molar. We have re-examined one of the syntypes of this species and can confirm the presence of a molar on the type material (see Figure 36).  Depth range. 335-390 m (Ledoyer 1986).
Discussion. Ledoyer (1986) originally described this species, tentatively placing it in the genus Bathycallisoma based on the relative length of the gnathopod 1 carpus, which is shorter than the propodus. We do not consider this to be a sound diagnostic character and instead refer to the distally broadened lacinia mobilis (slender robust seta in Bathycallisoma).   Ecology. Living over mud with sand.

Scopelocheiropsis sublitoralis G. Vinogradov, 2004
Discussion. Scopelocheiropsis sublitoralis is morphologically close to S. armata, both of which have a known distribution that is so far confined to Madagascar. Vinogradov (2004) does not justify his generic placement of the species, but presumably it was due to the absence of a molar, which S. abyssalis, the type of the genus, is now  (Ledoyer, 1986). Vinogradov, 2004. known to have. Nonetheless, S. sublitoralis exhibits characters which fit within the diagnosis of the genus.
Scopelocheiropsis sublittoralis can also easily be distinguished from S. abyssalis by the absence of a molar (present in S. sublitoralis); the shape of gnathopod 1, which is much more elongate and slender in S. abyssalis; and the shape of pereopod 7 basis, which has a long, thin posterodistal lobe in S. sublitoralis compared to the posteroventrally excavate corner of the pereopod 7 basis in S. abyssalis.
Etymology. The name Tayabasa refers to Tayabas Bay, located on the eastern side of Verde Island Passage in the Philippines, close to the type locality of the type species. Gender feminine.
Discussion. Tayabasa belongs to a closely related and highly derived complex of genera also comprised of Anisocallisoma, Austrocallisoma and Eucallisoma. See discussion under Austrocallisoma for further remarks. Tayabasa can be separated from Anisocallisoma by the 2-articulate maxilla 1 palp and inner plate with setae lining the inner margin (palp 1-articulate and inner plate with apical setae only in Anisocallisoma). It differs from Austrocallisoma in peg-like lacinia mobilis, the 2-articulate maxilla 1 palp, and the well-developed gnathopod 1 coxa (lacinia mobilis a stemmed, distally expanded blade, maxilla 1 palp 1-articulate, and gnathopod 1 coxa reduced in Austrocallisoma). Finally, it can be distinguished from Eucallisoma in the peg-like lacinia mobilis, the vestigial maxilliped palp article 4, and the subacute posteroventral lobe of the pereopod 4 coxa (lacinia mobilis a stemmed, distally expanded blade, maxilliped palp article 4 well developed, and pereopod 4 coxa with a subquadrate posteroventral lobe in Eucallisoma).  (Lowry & Stoddart, 1993).

Figure 39
Eucallisoma barnardi Lowry & Stoddart, 1993 Discussion. Eucallisoma barnardi is here transferred to its own genus, Tayabasa gen. n., on the basis of the cuspidate peg form of the lacinia mobilis, the vestigial maxilliped palp article 4, and the subacute posteroventral lobe on the pereopod 4 coxa.

Introduction
Research on marine molluscs of northern Chile began with the descriptions of some species by Sowerby (1832), d'Orbigny (1847), Hupé (1854) and Philippi (1860) in the late 19 th century. Further works include the studies done by Dall (1909), mostly in deep water areas along the Chilean and Peruvian coasts; Gigoux (1934), which listed the species found in the Region of Atacama, Marincovich (1973), describing the intertidal molluscs of Iquique; Acuña (1977), Bretos (1980), Bretos et al. (1983) and McLean (1984) dealing with fissurellid limpets and, more recently, the works of Guzmán et al. (1998) Plate (1901Plate ( , 1902, Thiele (1906Thiele ( , 1911 or Melvill and Standen (1912) among others -it was the work of Leloup (1956) which produced the most detailed study dealing with this molluscan class in the country; encompassing material from between Iquique (20°S) in northern Chile to Punta Arenas (53°S) in the extreme south of the country. Most of the subsequent works dealing with chitons have been focused on species from central and southern Chile (Castellanos 1948, 1951, Stuardo 1959, Osorio and Reid 2004, Schwabe and Sellanes 2004, Sirenko 2006, with a few works including shallow water species (Schwabe et al. 2006, Sirenko 2007. Further recent works including species from the Region of Atacama or northern Chile in particular only include the Kaas and Van Belle monograph series (Kaas and Van Belle 1985a, 1985b, 1987, 1990, 1994 which studied the worldwide chitons and, among them, Chilean species. This molluscan class is often overlooked in Chile, as their species are mostly small and hard to collect and to preserve, however, three of the large species of Chilean chitons (Acanthopleura echinata (Barnes, 1824), Chiton (Chiton) magnificus Deshayes, 1827 and Chiton (Chiton) granosus (Frembly, 1827)) are collected with gastronomic purposes (Osorio 2002), and some species are used in traditional medicine in northern Chile. Chitons play a role in controlling the green algal cover in mid-intertidal exposed rocky-shores of central Chile (Aguilera and Navarrete 2007) eating encrusting corallines (Camus et al. 2012), newly settled barnacles (Aguilera 2005) and other sessile and mobile invertebrates , and thus they have a direct impact on the intertidal ecosystem.
The coast of the Region of Atacama consists of rocky formations of volcanic origin with a few sandy beaches. The intertidal area of most of the coast, with the exception of a few scattered bays, is narrow (up to 20 m) and presents a diverse geography including cliffs, rocky plat- forms, intertidal pools, and boulder fields. The exposed side of rocks and boulders are exposed to strong surf, with just a few sheltered areas, particularly in the area of the Bay of Caldera, Obispito and Calderilla (Table 1). This work presents an overview, with distributions and illustrations, of all the species of Polyplacophora found in the Region of Atacama, northern Chile. The distribution range and a taxonomic key to all the studied species is also provided. The aim of this preliminary paper is thus to contribute on the knowledge of the molluscan fauna, in particular of northern Chile.

Material and methods
The material considered in this study was mostly obtained by sampling in the rocky coastal areas around the port of Caldera (27°04'S; 70°50'W), between Caleta Obispito (26°45'49"S; 70°45'17"W) and Puerto Viejo (27°20'23"S; 70°56'46"W), and in specific localities along the coasts of the Region of Atacama, northern Chile, during the summers of 2011 to 2012 and in August-December 2012. A synopsis of all the stations is given in Table 1 fig. 1. Detailed bibliography and synonymy in Kaas and Van Belle (1994: 161).
Description. Animal of small size, up to 11.5 mm in examined specimens, elongate oval, moderately elevated, color of tegmentum creamy white or bright white. Head valve semicircular, sculpture with 10-15 low rounded, equally spaced, nodulose, annulate, radial ribs, becoming obsolete towards the apex, the posterior ribs being strongest and more nodulose. Intermediate valves rectangular, lateral areas well defined, sculptured with two strong radial ribs. Central area with two series of equally spaced, diagonal lirae, forming rounded depressions in the interspaces. Tail valve semi-oval, slightly less wide than head valve, sculptured like head valve, with 8-11 weaker radial ribs. Girdle rather wide, yellowish white, dorsally covered with small, oval, imbricating scales (After Kaas and Van Belle 1994).

Material examined.
Specimens found under sunken rocks in tidal pools in Playa El Pulpito (SBMNH 452240, 1 speci men), Sur de Playa Ramada and Playa El Pulpo (Table 1).
Remarks. This is a small (under 12 mm) chiton, easily overlooked but for its bright whitish color. It is a fairly rare species; they were not abundant and were found only in two of the studied locations (Table 1), under rocks sunken in tidal pools, associated to small communities of Acar pusilla Sowerby, 1832, Liotia cancellata Gray, 1842, Rissoina inca Sowerby, 1832, encrusting algae and sponges. It has been reported that this species feeds on sponges, bryozoa and diatoms (Aguilera 2005). This species can be mistaken for Calloplax vivipara (Plate, 1902), differing from this species in having a less elongate body shape, with a much finer and subtle sculpture (especially noticeable on the terminal valves), it differs from C. vivipara in having rounded depressions in the central-lateral areas of the valves, especially in the middle valves.

Calloplax vivipara (Plate, 1899)
Plate 1, Fig. 3;  Ferreira (1978) as C. viviparus extended the southern distribution of this species at Valparaiso (33°02'S; 71°38'W). This is the first time this species is recorded, in the field, since its description and the present record fills a gap in the distribution in Chile of this rare species.
Remarks. This is a rare species, found in only two of the locations under study; in both places this species was found under rocks sunken in tidal pools, associated to encrusting sponges and to communities of the small mussel Brachidontes granulata (Hanley, 1843). According to Plate (1899) this species is ovoviviparous; that cited author found about 15 embryos, some with seven shell valves, in the ovary of a single specimen. In fact, this is the only chiton species ever reported to be ovoviviparous (Pearse 1979). This species is somewhat similar to Callistochiton pulchellus (Gray, 1828), differing in the coarse sculpture (especially in the anterior valve, with fewer and stronger ribs), the presence of longitudinal riblets in the central areas, and the more yellowish body color.  fig. 2A.

Description.
Animal of medium to large size, up to 45 mm long in examined specimens. Oval to elongate oval, slightly depressed, color of tegmentum greenish-brown to dark brown. Head valve semicircular, front slope straight, posterior margin V-shaped. Intermediate valves broadly rectangular. Tail valve less than semicircular, almost as wide as head valve. Girdle profusely beset with very long (up to 10 mm) thick, coarse, corneous hairs, not only interspersed throughout girdle but in girdle bridges, protruding at sutures and extending over valves. Tegmentum pustulose sculptured with minute and neatly separated pustules, on the end valves the pustules are arranged in radiating rows. Body width/length, mean 0.66; height/ length, mean 0.22 (After Ferreira 1983).
Material examined. Specimens found in almost all locations, with the exception of Bahia Cisne, Puerto Viejo and Playa Rodillo up to Obispito (Table 1). Calderilla (MPC-CL 3072014C, 1 specimen).
Remarks. This species is easily distinguished from all other chitons in the region by the presence of hairs covering the valves and sutures. A similar species, Chaetopleura (Chaetopleura) benaventei Plate, 1899 is slightly smaller in size and differs in the overall coloration and in the absence of the corneous bristles. Chaetopleura hennahi (Gray, 1828) found between El Callao, Peru and Arica, Chile (Kaas and Van Belle 1987) has a wine-red to reddish brown tegmentum and lacks the blackish corneous hairs which protrude at the sutures in Chaetopleura peruviana.
1/3 of girdle), up to 1.5-2 mm long in specimens 50 mm long (larger in larger specimens), vaguely striate, usually eroded at upper edge, clearly separated from each other by area as wide as scale; on outer 1/5 of girdle, scales much smaller, shorter, dark brown, erect, spine-like; girdle surface completely covered otherwise with minute, dark brown, lanceolate spicules, up to 100 μm long, 25 μm thick. Girdle bridges, empty in middle third, but crowded with small, dark brown spiculoid elements (akin to those on girdle proper) in outer thirds (After Ferreira 1986).
Material examined. Specimens found exposed on large boulders in the surf-zone, in Playa Rodillo, Playa El Pulpito (MPCCL 3072014E, 1 specimen) and in Norte Bahía de Caldera.

Remarks.
With sizes up to 200 mm (Sanhueza et al. 2008), this is one of the largest polyplacophoran species in the country. It lives almost exclusively in exposed rocks or in the surf zone. Although this species is mostly herbivore, it has been described also as a generalist polyphagous consumer, and a potential omnivorous, (Sanhueza et al. 2008). A brown-colored variety of the limpet species, Scurria variabilis (Sowerby, 1839), lives on the valves of this species, having been found in all the specimens examined in this study. An unidentified barnacle was also observed on the valves of a few specimens. In some places (Rodillo beach, Obispito bay; Table 1) juvenile specimens can be found among crevices of large boulders. It was observed that this species is predated by the common gull, Larus dominicanus (Lichtenstein, 1823), an omnivore species that also predates on the intertidal large keyhole limpets of the genus Fissurella (Bahamondes & Castilla, 1986).

Chiton (Chiton) cumingsii Frembly, 1827
Plate 1, Fig. 5; Table 2 Chiton cumingsii Frembly, 1827: 198, suppl Remarks. This colorful species is the most common and abundant chiton distributed in the zone; found in almost all the locations. It is commonly found in accumulations of several individuals on the underside of rocks at low tide, crawling quickly to the dark if exposed to sunlight. Among the examined specimens, some of them had a pink/orange coloration when juveniles with some adult specimens retaining a uniform pink coloration. This species has been cited as an introduced species in Las Palmas Port, Canary Islands (28°06'N, 15°25'W), being one of the few alien polyplacophoran found in European waters (Arias and Anadón 2013).
Material examined. Specimens found in Playa El Pulpo, Norte de Bahía de Caldera, Playa Mansa and Sur de Bahía de Caldera, in crevices in large rocks, often in surf zone in large colonies. Playa el Pulpo (MPCCL 3072014D, 1 specimen).
Remarks. This species is somewhat uncommon in shallower waters; it is found mostly in crevices and in rocky outcrops, mostly in clustered distributions. Juvenile specimens are somewhat similar to the juvenile specimens of Chiton cumingsii differing in the paler coloration (of various shades of green), the granulation on the valves and in the overall wider body. It has been reported that this species feeds on barnacle cyprids (Moreno andJaramillo 1983, Aguilera 2005) and is preyed on by the common gull Larus dominicanus (Lichtenstein, 1823). It was observed also that some specimens had barnacles on the valves. This species is a physiological omnivore, having the digestive flexibility and enzymatic capacity to digest and assimilate animal preys (Camus et al. 2009) and it is also a commercially important species (Osorio 2002).  Bullock (1988: 163).

Chiton (Chiton) magnificus Deshayes, 1827
Description. Animal of large size, reaching 115 mm in examined specimens. Body dark bluish-grey, broad-oval, slightly carinated, rather flat. Valves flattened to moderately carinated. Anterior valve sligthtly convex, semicircular, with wide V-shaped to straight posterior margin unnotched in middle, with numerous radially arranged, shallow ribs; intermediate valves rectangular with slight-ly concave posterior margin at both sides of faintly protruding apex, lateral areas slightly elevated, sculptured with up to 5 radial ribs between a wider diagonal ridge and a very wide posterior rib; tail valve semicircular with an anterior mucro; post-mucronal area with same sculpture as head valve and lateral areas (After Schwabe et al. 2006). According to Osorio (2002), this species can reach a maximum length of 174 mm.
Material examined. Specimens found in Aguas Verdes, Sur de Playa Ramada and in Playa Chorrillos, in subtidal areas attached to large boulders.
Distribution. Bullock (1988) gives a distribution for this species from Isla San Lorenzo, Peru south to Bahía Tictoc (43°36'40"S; 72°57'15"W), Chiloé Province, southern Chile. This species can be found in rock pools and boulder fields with strong water exchange, from the intertidal down to a maximum of 30.5 m depth at the Comau Fjord (42°23'S; 72°27'W), Region of Aysén (Schwabe et al. 2006). Smith and Ferreira (1977) considered the records of this species from Galapagos Islands as erroneous.

Remarks.
A shiny, large and conspicuous chiton, this species has been overlooked in recent molluscan literature, being cited by Valdovinos (1999) as Chiton latus and by Osorio (2002) as Chiton magnificus boweni. It seems to be an uncommon species, restricted to specific localities along the coast of Chile. Juvenile specimens may be misidentified as Chiton cumingsii, differing from this species in having a wider and flatter body, with smoother sculpture and with bright blue spots on the valves, which are cream white to greenish yellow in color. Apparently, in northern Chile this species is found only in subtidal areas. This is a commercially important species (Osorio 2002).
Remarks. This species was uncommon in the zone under study; only a few specimens were found in the undersides of rocks at low tide. This species is clearly identified from the other species found in this work by having a wide, flat shell, with narrower valves and a pattern of clear and darker alternating bands in the girdle. It can be misidentified as Chiton granosus; differing from this species in the smaller size, the much smaller girdle scales, a much weak valve sculpture and in the alternating bands of the perinotum, absent in Chiton granosus.

Family Chitonidae Subfamily Toniciinae Pilsbry, 1893
Genus  5I; Schwabe et al. 2006: 12, figs 9, 18;Gordillo and Schwabe 2009: 267, fig. 2D. A more detailed synonymy can be found in Kaas and Van Belle (1998: 25 Distribution. According to Reid and Osorio (2000), this species distributes in Chile between 40°S and 54°S, at Tierra del Fuego and around the Falkland Islands. The species ranges in depth from the low eulittoral to a depth of 36 m (Schwabe et al. 2006). The record presented here (at 27°S) is now the northernmost record for this species.
Remarks. Two specimens of this species were found in a single location; among a community of Tonicia chilensis, from which it distinguishes in attaining larger sizes and in having a darker body and almost smooth valves with minute granulation at the sides. It is interesting to note also that the valves of the examined specimens were widely separated, almost as in Tonicia disjuncta (Frembly, 1827). The presence of this species extends considerably the northernmost record of this species in about 1040 km (from 40°S to 27°S).
Description. Animal of medium to large size, reaching 43 mm in examined specimens. Shell elongate-oval, not much elevated, the dorsal ridge rounded, side-slopes straightened. Color umber-brown at the sides, becoming chestnut in the middle, delicately and peculiarly speckled and blotched and streaked with buff or buff-white. Lateral areas hardly raised, but separated from the central areas by an obtuse diagonal ridge bearing a series of low tubercles, sometimes subobsolete; sculptured with subradiating rows of small granules, and showing a band of irregularly placed black eyes on the forward part. Central area of second valve having in the middle, a keel or a group of lirae; central areas of the other valves having a narrow smooth dorsal band with several longitudinal furrows on each side of it; and at the sides there are longitudinal diverging delicate rows of granules. End valves radially sub-granulate, and crowded with eye-spots subradially arranged (After Pilsbry 1893).
Material examined. Specimens found on rocks at low tide in Aguas Verdes, Sur de Playa Ramada, Playa Mansa (MPCCL 3072014F, 1 specimen), Sur Bahía de Caldera and in Calderilla.
Distribution. According to Reid and Osorio (2000) this species distributes in Chile and Peru between latitudes 12° and 54°S. It has a bathymetric range from 0-28 m (Schwabe et al. 2006).
Remarks. This species has large and colorful mantles and plates of variable shades, which are similar to the encrusting calcareous algae commonly found in the rocky coasts. Due to the high diversity of forms, several synonyms have been described (see Kaas and Van Belle 1998), and this species needs a complete revision according to Schwabe et al. (2006). This species was found only in the lower intertidal to the subtidal areas, especially in protected locations. Schwabe and Sellanes (2010) reported 41 species of chitons from Chilean waters. Our results from the Region of Atacama, with eleven species found, accounts for 27 % of those reported species. All of the species occurring in the area have distributions in the southeastern Pacific Ocean, from Peru to southern Chile, with Calloplax vivipara, Radsia barnesii and Tonicia atrata as the only species endemic to the Chilean coast. The polyplacophoran diversity of the Region of Atacama is thus comparable to that described for southern areas of Chile, including central Chile (Aldea and Valdovinos 2005), the Comau fjord (Schwabe et al. 2006) and the Estero Elefantes and Laguna San Rafael areas (Osorio and Reid 2004), from where a similar diversity of this molluscan class has been recorded (with eleven, nine and nine species, respectively). With the exception of Callistochiton pulchellus, all of the species found in the Region of Atacama also occur in central and southern Chile.

Discussion
All of the studied species can be classified in two main groups according to their habitat; species with a higher relative frequency on exposed areas include the very large species Acanthopleura echinata and Enoplochiton niger. The other group includes species associated with protected intertidal areas: Callistochiton pulchellus, Calloplax vivipara, Chiton cumingsii, Chiton granosus (mostly found in rock fissures and crevices) and Radsia barnesii. Chiton magnificus was found in the Region of Atacama only in subtidal areas, always near large boulders in holdfast communities of the giant kelp Lessonia nigrescens. Chiton magnificus, however, is much more common in central and southern Chile, being found mostly in intertidal areas (Osorio 2002). The small-bodied species, Callistochiton pulchellus and Calloplax vivipara, were always restricted to submerged rocks in the bottom of tidal pools. This last habitat may explain the feeding behaviour of these small species, living over encrusting sponges and calcareous algae. Regarding feeding preferences; most of the large species of chitons from the Region of Atacama should have generalist diets, however it is possible that the smaller species have more specific diets, for example feeding in sponges, algal species, diatoms or barnacles.
The new distribution records of Callistochiton pulchellus, Radsia barnesii and Tonicia atrata and the new record of Calloplax vivipara may reflect the lack of sampling in the Atacama region or in northern Chile in general, where the scientific studies on invertebrates are still lacking. In particular, Callistochiton pulchellus and Calloplax vivipara may be more widespread in the country; however their particular habitat (and small adult size) may prevent their sampling by traditional methods. Some of these species may have also been overlooked or misidentified as juvenile specimens of other chiton spe-cies (for example Radsia barnesii as Chiton granosus). The considerable new range extension of Tonicia atrata found in this study may reflect the current complicated, unresolved status of the genus Tonicia in Chile; the revision of some particular species or species-groups is thus imperative, considering the great diversity in the valve and mantle morphology, which has derived in a large synonymy for some species, for instance for Tonicia chilensis (Schwabe et al. 2006).
The absence of other species, for example those cited by Valdovinos (1999) in the last complete revision of the Chilean mollusks (which have been traditionally considered as having distribution records in northern Chile), including Acanthochitona hirudiniformis (Sowerby, 1832), Acanthopleura granulata (Gmelin, 1791), Chaetopleura benaventei Plate, 1902, Chaetopleura hennahi (Gray, 1828, Ischnochiton imitator (Smith, 1881), Ischnochiton (Ischnochiton) punctulatissimus (Sowerby in Broderip & Sowerby, 1832) or Ischnochiton pusio (Sowerby, 1832) may be explained because the sampling activities in the Region of Atacama were restricted to, at most, sublittoral areas (2 m depth). Deep water areas must definitely harbor more unrecorded or undescribed species, as is the case with other invertebrate groups like sponges (Reiswig and Araya 2014) or stony corals (Araya et al. in prep.). It is, then, very probable that the number of chitons known from the Region of Atacama, or northern Chile in general, will increase with larger collecting efforts, including additional sampling methods such as dredges and samples from greater depths, even subtidal waters (incorporating also the bycatch of the commercial deep water fisheries). Like other zones of Chile, the deep water areas off Caldera (or off northern Chile in general) have not been investigated in detail and could yield interesting results.

Identification key
This key is primarily macroscopic (intended for identification of adult specimens) including external characters as shell features and general girdle features.

Introduction
Few works have been published on amphipods from Singapore and most are a century old. These works include Stebbing (1887), Mayer (1903) and Tattersall (1922). The genus Parelasmopus Stebbing, 1888 belongs to the family Maeridae Krapp-Schickel, 2008, with species typically occuring in the Indo-West Pacific ). To date only three species, Parelasmopus suluensis (Dana, 1853), P. setiger Chevreux, 1901 andP. dancaui Ortiz &Lalana (1997) are recognized with certainty from the Southeast Asian waters. Australia and the adjacent waters have the largest number of recorded taxa within this genus with seven species altogether namely, Parelasmopus ratory, amphipod specimens were sorted and preserved in 70% alcohol. The specimens were examined under a compound microscope and later selected for dissection. Specimens were introduced into increasing concentrations of glycerol before dissection was carried out in an excavated glass block with glycerol as a medium. Dissected parts were then permanently mounted in 100% glycerol. Dissections and mounting of specimens were carried out in glycerol. The appendages of the dissected specimens were examined using an Olympus SZ30 and figures were produced using an Olympus CH20 Leica light microscope with a camera lucida. All illustrations were digitally 'inked ' following Coleman (2003). Setae and mouthparts are classified following Watling (1989). The following abbreviations are used: A, antenna; G, gnathopod; HD, head; LL, lower lip; MD, mandible; MX, maxilla; MP, maxilliped; P, pereopod; EP, epimeron; T, telson; U, uropod; UR, urosome; UL, upper lip; R, right; L, left; ♂, male; ♀, female. All material is lodged with the Universiti Kebangsaan Malaysia Muzium Zoologi (UKMMZ).
Gnathopod 2 carpus relatively long about, 1.5 times as long as wide, slightly lobate; propodus linear, almost five times as long as broad, without distomedial shelf; dactylus apically subacute.
Remarks. Chevreux (1901) described Parelasmopus setiger from Port of Victoria, Mahé, Seychelles. His description was based on a male of 7 mm body length, with figures of a lateral view, mandible, maxilliped, accessory flagellum of antenna 1, gnathopods 1 and 2, uropod 3 and telson. In , he recorded the occurrence of P. setiger from the Philippine Islands and discussed the possible occurrence of P. setiger that include tropical Australia, Indonesia and the northern Indian Ocean. He also pointed out the difficulty in working with the Indo-Pacific Parelasmopus, due to growth stage and historic identification being mixed. Barnard identifies P. albidus, P. setiger, P. suluensis and P. suensis in particular as requiring revision before further progress can be made. Hughes (2011) identification of P. cf. suensis restates this problem (p77).
Both Chevreux's (1901) and  figures of the species are similar to ours, antenna 1 peduncle with 2 setae, male gnathopod 2 propodus palm transverse with posteroproximal elevation with 6 robust setae, pereopods 5 to 7 posterior margins with long slender setae and the dorsal carina pattern for pereonite 7 and pleonites 1 to 3. Thus, our male specimen agrees well with the original description of Chevreux (1901), except for a few minor differences, such as the serrated robust setae on the inner surface of the outer plate of maxilla 1. However, this could possibly be the next developing stage of the new growth (internal growth of next instar) of a maxilla 1. Additionally, our female specimens are observed without paired dorsal carina in pereonite 7.
The recently described P. siamensis Wongkamhaeng et al. (2013) has close resemblance to our specimens. As both P. siamensis and our specimen is considered to be in their terminal adult stage, they share the same form of the gnathopod 2 palmar margin with posteroproximal elevation with 6-7 robust setae, the midposterior toothed dactylus and the serration on coxa 1-3. Therefore P. siamensis Wongkamhaeng et al (2013) is here synonymized with Parelasmopus setiger Chevreux (1901). Until now the species was recorded from Seychelles, Philippine Islands, Sulu Sea, Indonesia, Australia, north Indian Ocean, Gulf of Thailand and Singapore. The present records confirm this distribution.
The present comparison suggests that further taxonomic studies on this species group are necessary. Detailed drawings and descriptions provided in this study could aid in eliminating further confusion within the P. setiger complex, including and thus establish its definitive characteristics.
Distribution. Seychelles, Philippine Islands, Sulu Sea, Indonesia, Gulf of Thailand, Australia, north Indian Ocean and Singapore (current study).

Introduction
The colubrid genus Oligodon Fitzinger, 1826 is currently known to include 75 valid species (Uetz andHallermann 2014, 1 st September 2014). Only four species have been recorded from Sri Lanka: Oligodon calamarius (Linnaeus, 1758); Oligodon arnensis (Shaw, 1802); Oligodon taeniolatus (Jerdon, 1853); and Oligodon sublineatus Duméril, Bibron & Duméril, 1854. Following the description of Oligodon sublineatus in 1854, this species has since been recorded from various locations in Sri Lanka (Boulenger 1890(Boulenger , 1894Wall 1921;Smith 1943; until now, was not recognized as a syntype. However the smaller specimen (MNHN 3239) has been mistakenly considered as the holotype by Wallach et al. (2014). We have identified the large specimen as being one of the syntypes, so we hereby designate it as a lectotype, and redescribe it in detail in order to stabilize that name with a recognised type specimen.

Methods
Museum acronyms follow Sabaj Pérez (2014). Specimens were examined in the collections of the British Museum of Natural History, UK (BMNH); Muséum national d'Histoire naturelle, France (MNHN); Naturhistorisches Museum Basel, Switzerland (NMB); and National Museum of Sri Lanka (NMSL). Morphometric and meristic data for species comparisons were obtained from examined specimens (see Appendix 1). We checked the external morphology of specimens with a Wild M3Z stereomicroscope and photographed them with a Canon EOS 7D SLR digital camera. The map was constructed based on Cooray (1967). The conservation status of the species was evaluated using Red List Categories and Criteria in IUCN Standards and Petitions Subcommittee (2013: version 10.1) to assess their risk of extinction. Sex was determined by ventral tail incision of adult specimens followed by the checking for the presence or absence of hemipenes. All the natural history data were taken from our own field observation notes made during the last ten years.
The following characters were measured with a digital caliper (±0.1 mm) on the left side of the body for symmetrical characters: eye diameter (ED, horizontal diameter of eye); eye-nostril length (EN, distance between anterior most point of eye and middle of nostril); snout length (ES, distance between anterior most point of eye and snout); nostril diameter (ND, horizontal diameter of nostril); internarial distance (IN, least distance between nostrils); mandible-posterior eye distance (MPE, distance between posterior edge of mandible and posterior most edge of eye); interorbital width (IO, least distance between upper margins of orbits); head length (HL, distance between posterior edge of mandible and tip of snout); head width (HW, maximum width of head); snout-vent length (SVL, measured from tip of snout to anterior margin of vent); tail length (TAL, measured from anterior margin of vent to tail tip). Meristic characters were taken as follows: supralabials and infralabials (SUP and INF, first labial scale to last labial scale bordering gape); costal scales (COS, counted around the body from one side of ventrals to the other in three positions, on one head length behind neck, at mid body and at one ventral scale prior to preanal); when counting the number of ventral scales (MVS), we scored specimens according to method described by Dowling (1951). We counted subcaudal scales (SUB) from first postcloacal scale to the scale before the tip of the tail.

Oligodon sublineatus Duméril, Bibron & Duméril, 1854
Figs 1-3; Tables 1, 2 Remarks. Standard morphometric and meristic data of the two syntypes are presented in Table 1. We hereby recognise two syntypes: the larger specimen (MNHN 3238) and the smaller specimen (MNHN 3239). Uncertainties still exist in Oligodon taxonomy and O. sublineatus may represent a cryptic species complex in Sri Lanka (see table 2 showing the wide range of subcaudal and ventral counts within O. sublineatus), therefore it is necessary to stabilize the name with a recognised lectotype. There are two main reasons for selecting MNHN 3238 as the lectotype: (1) it was used in the original description and its morphometric data has been provided and (2) it is a fully grown, well-developed and well preserved adult specimen in good shape.
Lectotype (here designated). MNHN 3238, adult female collected from the Philippines (mistakenly so in the original description) [from Java (also in error) according to the museum registry] by an unknown collector [by Bosc (in error) according to the museum registry].

Diagnosis.
Oligodon sublineatus shows sexual dimorphism in scalation (Table 2) and is distinguished from all congeners by the following characters: SVL 152-310 mm; TAL 20.0-42.0 mm; 130-161 ventrals; 23-42 subcaudals (divided); anal plate divided; loreal present; seven supralabials; temporals 1+2; ventral side with three series of dark brown points forming almost continuous stripes, with the middle series of points absent on the tail; dorsal coloration (live or in alcohol) greyish brown, speckled with small elongated spots irregularly placed; posterior part of the jaws has a large, oblique spot extending along the neck posteriorly; dorsally a "˄" shaped marking between the eyes, which continues laterally across them; an irregular, brown, transversal band from the frontal to the post-parietal region.

English translation of the original French description in Duméril, Bibron & Duméril (1854: 57). Characters. Ventral side with three series of points forming stripes.
This species is mostly characteristic, as its specific name, by having three black stripes along the ventral side, which are made up of a series of points, meeting together. The two stripes outside the ventral plates form a continuous line up to the ventral surface of the tail, but the central one is made up of distinct points in the centre of the ventral plates. These points are quite large, round and wide posteriorly, and are as notched at the front; the median stripe does not prolongate onto the ventral side of the tail.
Dorsal coloration grey, speckled with lines or with small elongated spots irregularly placed; however, around the anterior third of the body and laterally, three of those spots appear enlarged with increased width, having a circular border. The spots are constricted central-ly and have white borders. The posterior section of the jaws has a large, oblique patch along the neck posteriorly where it forms a tip pointing in the opposite direction to the characteristic collar of the first species [note from the translator: Oligodon sub-quadratum].
Dorsal scales are very smooth, and are close to each other; they are slightly overlapping, like roof tiles, mostly around the tail area, and in this respect, very skink-like in appearance.
Rostral plate is notched, and crescent shaped; other plates covering the head are large and clearly distinct as in colubrids.
We were only able to examine one well preserved specimen, having no clues as to the origin of the specimen [the Philippines] and the name 'Oligodon torquatus' appears along with the letter "R" on the jar.
Another specimen, younger and obviously added much later, had a median stripe made up of numerous spots which were less distinct, was collected from Ceylan by Mr. Leschenault. This specimen bears all the characters previously described: the large, brown, post-maxillary mark set posteriorly on the neck forming a croissant shape; with a laterally set, black mark extending onto the anterior third of the body.
Total length was 180 cm [sic]; among them 155 for SVL and 25 for the tail. Rostral shield large, hemispherical, distinctly visible from above, pointed posteriorly; interorbital width broad (IO 78.7% of HW); internasals semicircular; nostrils rather large; nasals completely divided by nostrils into two scales unequal in size; anterior nasal larger, in anterior contact with rostral, internasal dorsally, 1 st SUP ventrally; posterior nasal in contact with internasal and prefrontal dorsally, loreal posteriorly, 1 st and 2 nd SUP ventrally; prefrontal rather large, broader than long, and subhexagonal; frontal large, subhexagonal, elongate posteriorly and longer than its width; supraoculars narrow, elongated, subrectangular, posteriorly wider; parietals large, butterfly wing-like in shape, bordered by supraoculars, frontal, upper postoculars anteriorly, anterior and upper posterior temporals, and six dorso-nuchal scales posteriorly; loreal large, slightly elongated, subrectangular, in contact with prefrontal dorsally and preoculars posteriorly, ventrally only touching the 2 nd SUP; one preocular (both sides), vertically elongated, subrectangular, in contact with prefrontal and loreal anteriorly, supraocular dorsally, and 3 rd SUP ventrally; eye moderate (ED 15.7% of HL), ellip-  tical, nearly a half of the size of snout length (ED 50% of ES), pupil rounded; two postoculars, upper postocular smaller, quadrangular, contact with supraocular and parietal broad, in narrow contact with anterior temporal; lower postocular crescent in contact with 4 th and 5 th SUP ventrally, anterior temporal posteriorly; temporals 1+2, elongated, hexagonal; anterior temporal larger and longer than posterior temporals, in contact with parietal dorsally, 5 th and 6 th SUP ventrally; posterior temporals smaller, lower one in contact with 6 th and 7 th supralabials ventrally. Supralabials 7 (on both sides), 4 th -7 th larger in size; 1 st SUP in contact with rostral anteriorly, nasals dorsally, 2 nd supralabial with posterior nasal and loreal dorsally, 3 rd SUP with preocular and orbit dorsally, 4 th SUP with orbit and the lower postocular dorsally, 5 th SUP with lower postocular and anterior temporal dorsally, 6 th supralabial with anterior temporal and lower posterior temporal dorsally, and 7 th SUP with lower posterior temporal dorsally and body scales posteriorly.
Mental of moderate size, triangular; first infralabial pair larger than mental plate and in broad contact with each other, in contact with anterior chin shield posteriorly; eight infralabials, 1 st -5 th in contact with first chin shield, 5 th infralabial largest in size in narrow contact with the anterior chin shield and in broader contact with the posterior chin shield; 6 th -8 th infralabials in contact with gular scales; two larger anterior chin shields, and two smaller posterior chinshields all in broad contact; posterior chin shield bordered posteriorly by six gular scales.
Rostral shield large, hemispherical, distinctly visible from above, pointed posteriorly; interorbital width broader, IO 80.5% of HW; internasals semicircular; nostrils rather large; nasals divided into two scales unequal in size; anterior nasal larger, in contact with the rostral plate anteriorly, internasal dorsally, 1 st SUP ventrally; posterior nasal in contact with internasal and prefrontal dorsally, loreal posteriorly, 1 st and 2 nd SUP ventrally; prefrontal rather large, broad, and subhexagonal; frontal large, subhexagonal, elongate posteriorly and longer than its width; supraoculars narrow, elongated, subrectangular, posteriorly wider; parietals large, butterfly-like in shape, bordered by supraoculars, frontal, upper postoculars anteriorly, anterior and upper posterior temporals, and six dorso-nuchal scales posteriorly; loreal large, slightly elongated, subrectangular, in contact with prefrontal dorsally, preoculars posteriorly, posterior nasal anteriorly, ventrally just meets the 2 nd SUP; one preocular in both sides, vertically elongated, subrectangular, in contact with prefrontal and loreal anteriorly, supraocular dorsally, and 3 rd SUP ventrally; eye moderate, ED 17.3 (17.6)% of HL, elliptical, nearly a quarter of the snout length, ED 51.9 (56.2)% of ES, pupil rounded; two postoculars, upper postocular smaller, quadrangular, in contact with supraocular and parietal broad, in narrow contact with anterior temporal; lower postocular crescent in contact with 4 th and 5 th SUP ventrally, anterior temporal posteriorly; temporals 1+2, elongated, hexagonal; anterior temporal larger and longer than posterior temporals, in contact with parietal dorsally, 5 th and 6 th SUP ventrally; posterior temporals smaller, lower one in contact with 6 th and 7 th SUP ventrally. Supralabials 7 on both sides, 4 th -7 th larger in size; 1 st SUP in contact with rostral anteriorly, nasals dorsally, 2 nd SUP with posterior nasal and loreal dorsally, 3 rd SUP with preocular and orbit dorsally, 4 th SUP with orbit and the lower postocular dorsally, 5 th SUP with lower postocular and anterior temporal dorsally, 6 th SUP with anterior temporal and lower posterior temporal, and 7 th SUP with lower posterior temporal dorsally and body scales posteriorly.
Mental moderate, triangular; first infralabial pair larger than mental and contact with each other broad, in contact with anterior chin shield posteriorly; eight infralabials, 1 st -5 th in contact with first chin shield, 5 th infralabial largest in size in narrow contact with anterior chin shield and contact with posterior chin shield broad; 6 th -8 th infralabials in contact with gular scales; two larger anterior chin shields, and two smaller posterior chinshields all in broad contact; posterior chin shield bordered posteriorly by six gular scales.
The result of the application of the IUCN (2013) B2 a, b (iii) Red List criteria shows that O. sublineatus as Least Concern (LC): recorded from an altitude range of 10-1600 m in all vegetation zones of Sri Lanka. Its area of occupancy is 6,000 km 2 , and its extent of occurrence is 40,000 km 2 . Natural history. A nocturnal snake, sometimes active during day time. Temperature, humidity, and light intensities for daytime activity were respectively measured at 24.8-27.2 °C, 67-82%, and 38-365 lux, based on 50 observations in dense forested areas. It usually does not bite, but if this does occur then it will lead to soreness, pain and temporary bleeding in the victim. Biting has been occasionally observed during touching or handling attempts by the victim. When frightened, the snake either coils up and hides its head within its coiled up body; or it quickly tries to escape to a safe hiding place inside the leaf litter.
When the snake coils, it enlarges its body and displays its vivid skin colours (white, pink and brown), which is visible between the scales around the mid body. We observed, on a number of occasions, the snake practicing thanatosis (death mimicry) for up to 10-15 minutes after carrying out our own handling attempts. Once the snake had noticed that threat had disappeared, it quickly escaped and hid itself in the leaf litter. We have observed this species living in sympatry with other snakes of several families such as Aspidura guentheri Ferguson, 1876 (Natricidae); Hypnale zara (Gray, 1849) (Viperidae); and Sibynophis subpunctatus (Duméril, Bibron & Duméril, 1854) Based on our observations, its diet consists mostly of lizards (saurophagy) and small snakes eggs (oophagy), small spiders, beetles, other insects and the larvae of other invertebrates. More specifically, we observed the snake feeding on ground dwelling skinks (Lankascincus sp.) and geckos (Hemidactylus frenatus and Cnemaspis sp.). If the prey is large, the snake wraps itself around it and squeezes it until it suffocates. In captivity, it was fed with jumping spiders, small wild cockroaches, annelid worms, meal-worms, small frogs, and the freshly detached tail tips of geckos.
During the breeding season (May-June) 3-5 individuals can be observed close by and we observed several copulations in the evenings just after dark (18.0-19.0 hrs). The species lays 3-5 eggs at a time on dry, cool, loose soil or under decaying logs on the ground (soil temperature 26.2-27.9 °C; humidity 58-73%; light intensity 0-27 lux, based on observations of 10 ovipositions). Eggs are cream in colour and oval in shape (12-14 mm long and 4-5 mm wide, n = 40). The lectotype MNHN 3238 is a gravid female with three eggs in its genital tract. The incubation period is 38-45 days (based on observations of 10 incubating clutches). We did not see the parents close by during the incubation nor shortly afterwards, indicating the lack of parental care of the eggs or hatchlings. The new born juveniles were 4-5 cm in total length and their body colour varied from dark brown to black. We noticed that ants were their main egg predators on about ten occasions. We also observed on several occasions, this snake attempting to avoid ant-nests when moving or resting.
We have found this species inside termite mounds on many occasions, an observation also made by Smith (1943). This may indicate either a strategy used by the snake to avoid ants (because we never observed ant nests in or around termite mounts) or a neat way for the snake to have instant access to food (may be feeding on termite eggs). Further studies on habitat ecology would be interesting. Even though this is a ground dwelling species, we observed it climbing on rock boulders which have crevices, indicating that this snake may be searching for geckos or their eggs for food. During floods, the snake is usually found off the floor, in trees at 1-2 m above ground level. It is also found deep inside forests, and has been observed under old coconut harnesses, decaying logs on the ground, and inside termite mounds (as mentioned earlier) set in well maintained home gardens.

Discussion
In the description of Oligodon sublineatus, Duméril et al. (1854) clearly states the following "We only have ob- served one specimen well preserved …..", they further stated "We counted 15 scale rows on that specimen, 155 ventrals and 25 subcaudals". Those counts are in accordance with MNHN 3238 (respectively 150 and 28) hence; we hereby designate it as the lectotype. However, the measurements given in the last line "Total length was 180 cm; among them 155 for SVL and 25 for the tail." is a mistake; we believe that the wrong units of measurements were chosen in error; it should have been in millimeters and not centimeters! In addition, the newly recognized syntype (MNHN 3238) had a total length of about 289 mm with 254 mm SVL and 35 mm for the tail. Again we are making the assumption that Duméril et al. (1854) must have mistakenly typed the total length as "180cm" instead of ~280 mm and "SVL 155cm" instead of ~255 mm (typing a '1' instead of a '2'). Interestingly, the syntype MNHN 3239 (now paralectotype) measured 177 mm total length with SVL 150 mm but its ventral and subcaudal counts do not match those of the original description (respectively 138 and 36 [typical of a male] versus 155 and 25 in the description [typical of a female]). However, the most probable explanation of this is that they mistyped, rather than used (which may seem the obvious explanation here) the measurement of the second specimen, because Duméril et al. (1854) clearly stated that they had examined only one specimen (the largest of both syntypes), even though they compared the colour patterns of both specimens, thus both are here considered as syntypes. Furthermore, the scale counts in the smaller specimen (MNHN 3239) do not match the original scale description of Duméril et al. (1854), and the spots of the larger specimen (MNHN 3238) are much more narrowed towards the middle of the body compared to the spots of the smaller specimen (MNHN 3239), which is in accordance with the details of the examined specimen in the original description. Thus Duméril et al. (1854) made a mistake when describing characteristics of the examined specimens and their ventral and subcaudal counts also reflect the classical mistake often seen when one single, old and very small specimen is examined by many different researchers over time. We have no doubt that MNHN 3238 is the Philippines (in error) specimen of the original description and MNHN 3239 the Ceylan specimen of Leschenault as reported in the original description, both being the only two syntypes of O. sublineatus.
To be sure that there are no other specimens which could possibly be a syntype, we examined all the available Oligodon sublineatus specimens and all the specimens of Oligodon collected from Sri Lanka and the Philippines which were registered on or before 1864 in the MNHN collection (1864 is the date of the oldest handwritten register available for the MNHN herpetological collections). Among the available specimens (except MNHN 3238-39), MNHN 0611, 3537, 5768, 1900MNHN 0611, 3537, 5768, .0381-385, and 1900MNHN 0611, 3537, 5768, .0381, 1900 Leviton (1963)]. Furthermore, based on the description, it can also be easily distinguished from O. torquatus (Boulenger, 1888) by having ventrals with a series of dark brown spots in three lines (vs. uniform).
Boulenger (1894) recorded Oligodon sublineatus from Nicobar Island. Deepak and Harikrishnan (2013) observed a couple of specimens (ZSI 8899 and 8900) of O. sublineatus deposited at ZSI-Kolkata, which were labelled as "Camorta, Nicobars". They confirmed that both the collection locality and the identity were wrong. The species is definitively absent from Nicobar Islands, as originally stated by Wall (1921) and has to be considered as a species strictly endemic to Sri Lanka, but widespread over the forested areas of the country.

Introduction
In a study that appeared in the journal "Nature" on the 6 th of April 1882, just a few days before his death, Charles Darwin returned to his lifelong fascination with mechanisms of passive long-distance dispersal in mollusks (Darwin 1882). Ever since Darwin's pioneer work, land snails occurring on islands have puzzled evolutionary biologists; it is indeed challenging to explain how these fragile and slow moving creatures could travel long distances across unsuitable ecological areas if not by passive  (Ronquist et al. 2011). For both analyses the most appropriate model of sequence evolution was selected using JMODELTEST 2 (Darriba et al. 2012). JMODELTEST 2 returned the HKY+I+G as the best model fitting the concatenated dataset (proportion of invariable sites I=0.575; Gamma distribution shape parameter = 0.576). These settings were then adopted in the Maximum Likelihood, Bayesian and Beast searches. The robustness of the ML hypothesis was tested by 1,000 bootstrap replicates; MrBAYES was run two times independently for 2,000,000 generations with a sampling frequency of 100 generations. We ran one cold and three heated Markov chains and two independent runs. To establish if the Markov chains had reached stationarity, we plotted the likelihood scores of the sampled trees against generation time. Trees generated before stationarity were discarded as burn-in (first 10% of the sampled trees) and posterior probability values for each node were calculated on the basis of the remaining 90% of sampled trees. We applied coalescence as implemented in the BEAST 1.7.2 package (Drummond and Rambaut 2007) to estimate divergence times in million years (Myr) for the supported clades found by the phylogenetic analyses. BEAST was used to estimate node ages to the most common recent ancestor (TMCRA) of the splits and substitution rates using an uncorrelated lognormal relaxed clock with a Yule or 'pure birth' prior process to model speciation. The output of each independent run was visualized using TRACER v1.4. Samples from both independent runs were then pooled after removing the first 10% as burn-in using LogCombiner 1.4.8. After an optimization step during which parameters were calculated to reach an optimum performance and achieve a reasonable effective sampling size (ESS, number of independent samples of the posterior distribution) for the parameters of interest, we carried out two independent runs of 30 million generations each, using a Yule tree prior and the default options for all other prior and operator settings. The age of the basal split in the geographically and phylogenetically closely related genus Codringtonia, estimated at 4.4 Myr in Kotsakiozi et al. (2012), was used to calibrate the tree with a normal distribution for the prior. Finally, we conducted a Bayesian binary MCMC dispersal-vicariance analysis in RASP (Yu et al. 2011). Analyses were run with the maximum units of areas allowed in ancestral nodes equal to three; other parameters were kept at the default settings.

Results
The final alignment including all samples amplified by nested PCR was 482 base pair (bp) long, with 281 bp for COI and 201 for 16S and defined a total of 33 unique haplotypes (GenBank accession numbers KR080942-KR081055; additional sequences used for the phylogenetic searches are JQ239955, JQ239934, JQ239967, JQ239977, JQ240123, JQ240103, JQ240134, JQ240145, HQ203051 and EU912763). These haplotypes are robustly clustered in a monophyletic clade in all phylogenetic searches (Fig. 1). The two Assyriella taxa, viz. guttata and mardinensis, and the Levantina clade were found not to be reciprocally monophyletic and the placement of Assyriella is not supported statistically.
Levantina hierosolyma, L. caesareana and the two L. spiriplana forms from the Dodecanese traditionally considered as subspecies (i.e. spiriplana spiriplana and spiriplana malziana) (Glaubrecht 1993a(Glaubrecht , 1993b were not retrieved as separate clades. This claims for a revision of the systematics of the group, which is outside the scopes of this study and would require a denser sampling in terms of taxa, geographic coverage and molecular markers. Such a larger study would also allow alternative hypotheses such as incomplete lineage sorting and introgression to be tested in details. We found strong support for three main haplogroups, in spite of the short mtDNA fragments we were able to amplify and sequence. Haplotypes 12, 16, 31 and 33 are from non-umbilicate shelled populations distributed in Turkey, Israel as well as on Symi and their last common ancestor is not older than about 3 Myr old. A large group not older than 3 Myr clusters populations with umbilicate shells distributed over a broad geographic area from Israel, Turkey and the Dodecanese (Karpathos, Rhodes and Symi Islands); haplotypes 23 and 27 are present on Rhodes and Symi Islands and on Rhodes and in Israel, respectively. The remaining haplotypes are gathered in a clade with non-umbilicate shells only; this cluster is limited to the Dodecanese and dates back to about 2.6 Myr ago. Within this clade, in three circumstances the same haplotype is shared between two islands. The RASP analysis (Fig. 2) postulates an origin of the Levantina / Asyriella clade in Turkey (node I; occurrence of this range 92.2%; 80% marginal probability) followed by two dispersal events to the Dodecanese (node II) and the Levant (node IV), respectively. Node V grouping umbilicate populations implies a back dispersal from the Levant to the island of Rhodes followed by multiple dispersal events across Rhodes, Karpathos and Symi. The distribution of haplotypes 19, 27 and 30 supports a close association of umbilicate forms in the Levant, continental Turkey and Rhodes where the umbilicate form (spiriplana) exclusively occurs in the vicinity of the Crusaders' fortresses (node VII; 82% marginal probability). The lineage with a covered umbilicus (malziana) originated on Rhodes (node VIII; occurrence of this range 93.7%; 79% marginal probability) and progressively spread and diversified across Rhodes, Karpathos and Symi. Haplotypes found on Symi and on the neighboring islet of Nimos (haplotypes 13, 14, 24 and 15) reached the islands via two independent dispersal events from Rhodes (nodes XI and XIII).

Discussion
We could not retrieve two reciprocally monophyletic clades for Levantina and Assyriella. The validity of the distinction between Assyriella and Levantina was already questioned (Glaubrecht 1995); this is an issue in need of further work based on a taxonomically more exhaustive sampling. Our results do not fully embrace neither the paleogeographic nor the Crusaders hypothesis but rather suggest that both played a role in shaping the observed mtDNA diversity in Levantina. The umbilicate and non-umbilicate shell forms, although clearly distinguishable morphologically (Glaubrecht 1993b(Glaubrecht , 1995, do not cluster in two reciprocally monophyletic lineages. #The umbilicate populations are restricted to a single and highly supported monophyletic clade whereas the non-umbilicate shell is displaced as a paraphyletic trait in the tree of Fig. 1. It is also worth emphasizing that in our phylogeny only the non-umbilicate shell form is associated to both basal and terminal branches. The biogeographic reconstruction presented in Fig. 2 supports multiple long-distance and over-sea dispersal events; some of these events are phylogenetically very recent (Figs 1 and 2). The scenario we detail here requires particular caution in its interpretation for the following reasons. First, inferences are based on mtDNA only, a single genetic locus and hence do not give any indication on the geographic origin of the rest of the genome. Second, the uncertainty of the tree topology is not taken into account. Nonetheless, in our opinion it represents a likely scenario given the phylogenetic and geographic information available and the ecology of the group. The origin of the whole Levantina clade, with the inclusion of one of the "Assyriella" taxa, should be sought in continental Turkey (node I; Fig. 2). Nodes V and VIII are roughly coeval (Fig. 2; 3.03 -2.64 Myr, ages in bold in Fig. 1) and group only insular non-umbilicate populations (malziana) on one hand and almost exclusively insular umbilicate populations (spiriplana) on the other. The exceptions to insularity in the umbilicate clade will be discussed later on; we anticipate that these are, in our opinion, due to anthropogenic translocations. The above time estimates perfectly overlap with the last connection of Karpathos and Rhodes to the mainland, which dates back to 3.5 -2.8 Myr ago (Beerli et al. 1996); this suggests that Levantina reached the Dodecanese exploiting continuity of landmasses. This, however, did not happen via a single colonization event but rather through a twofold process, which brought on Rhodes two immigrant lineages of independent origin. The reconstruction of Fig. 2 identifies a first dispersal event from continental Turkey to Rhodes (nodes II), which resulted in the insular non-umbilicate clade widespread there (malziana). A second wave of colonization arrived on Rhodes following an earlier dispersal from Turkey to the Levant (nodes IV and V); this clade had the latter as the most likely area of origin and marked the appearance of the umbilicate shell type (spiriplana). Subsequently, both lineages colonized Karpathos and Symi (nodes VI-XIII); the non-umbilicate lineage realized more across-island movements (nodes VIII-XIII) than the umbilicate lineage (nodes VI-VII). How and when Levantina dispersed from Turkey to the Levant and why the umbilicate shell type is mostly insular are questions that remain to be answered. Recent evidence suggests that shell shape diversification in Mediter- Numbers at nodes are statistical support for the ML and Bayesian searches (first and second value above branches). Numbers below branches are age estimates in millions of years with the 95% highest posterior density (HPD) credibility interval in parentheses. Age estimates in bold are discussed in details in the text. Haplotype numbering is as in Table 1. The distribution of each haplotype and the relative shell shape is summarized in the column to the right of the haplotype identifiers (K = Karpathos Is.; R = Rhodes Is.; S = Symi Is.; N = Nimos Is.; CT = Continental Turkey; IS = Israel). Pictures illustrate how shell variability (closed or open umbilicus; squares and circles, respectively) is distributed in Levantina and Assyriella. ranean land snails might be due to the interplay between historical and ecological factors (Fiorentino et al. 2013). The positioning of haplotypes from the Levant at the base and within the otherwise poorly differentiated umbilicate clade suggests that the occurrences of this form in the Dodecanese and in southwestern Turkey is probably the result of (anthropogenic?) introductions. Such a shallow level of differentiation in the umbilicate clade is not mirrored in the non-umbilicate clade.
Even though the hypothesis of the non-umbilicate clade being of insular origin is appealing from an evolutionary perspective, we should not overlook the fact that this result could be an artifact due our limited sampling. In particular, we cannot completely rule out the hypothesis that the non-umbilicate clade originated in the Levant and subsequently reached the Dodecanese. We indeed identified a non-umbilicate clade grouping haplotypes from both the Levant and the Dodecanese (haplotypes 12, 16, 31, and 33). The non-umbilicate samples from the Levant are genetically close to a few samples from Symi. The other non-umbilicate individuals from that island are spread in the upper part of the tree of Fig. 1, scattered among haplotypes found on other islands. In order to test this hypothesis adequately we would have needed a better survey of the genetic variation in Levantina from the Levant that is, on the contrary, too limited to address the issue.
Rhodes, the largest island of the Dodecanese and the closest to the mainland, is identified as the first colonized by the two Levantina lineages (nodes V and VIII). The non-umbilicate form is widespread across the island while umbilicate-shelled individuals are restricted to the fortress of the Knights Hospitaller of St. John in the northern part of island (city of Rhodes) (Glaubrecht 1993b(Glaubrecht , 1995. This has been taken as an indication that humans introduced this shell type on the island (Glaubrecht 1993b(Glaubrecht , 1995. Our molecular time estimates reject this hypothesis, at least for the basal haplotypes involved in the event (haplotypes 11 and 22). In our phylogeny, however, we found five cases of haplotypes shared between islands or between islands and distant locations. These distri-  Fig. 1 but pruned of the outgroup taxa) summarizing the Bayesian dispersal -vicariance analysis. The distribution of each haplotype and the relative shell shape is summarized in the column to the right of the haplotype identifiers (K = Karpathos Is.; R = Rhodes Is.; S = Symi Is.; N = Nimos Is.; CT = Continental Turkey; IS = Israel). Pie charts and numbers next to them indicate marginal probabilities of alternative ancestral ranges; colors identify the different geographic areas considered and match those in Fig. 1. Roman numbers identify events discussed in the text. On the right is the schematic of the proposed biogeographic history of Levantina. Arrows indicate the direction of the dispersal events inferred by the Bayesian dispersal -vicariance analysis and discussed in the text; roman numbers are the same as in the cladogram shown on the left. The bottom left panel details events within the umbilicate clade (circles), the bottom right panel those within the insular non-umbilicate clade (squares).
butions are difficult to justify on natural bases; the lack of any genetic differentiation indicates very recent dispersal in spite of the intervening marine barriers. Within the umbilicate clade, node VII suggests a recent dispersal event across Rhodes, continental Turkey and Levant; the same haplotype 27 co-occurs on Rhodes and in the Levant. Similarly, haplotype 23 is present on Rhodes and Symi. Haplotypes 1, 2, 3 and 5 (all non-umbilicate) are shared between two islands (alternatively Karpathos and Rhodes or Karpathos and Symi); most interestingly, on Rhodes these haplotypes are confined to human settlements (Table 1). How could we explain the allopatric occurrence of the same (or slightly diverging) haplotypes currently separated by geographic barriers theoretically insurmountable by land snails? Over-sea passive dispersal through i.e. hitchhiking on birds and/or surviving the passage through birds' gut has been documented in land snails (Gittenberger et al. 2006;Miura et al. 2012). It should also considered that large helicids are edible and they have been found associated with human settlements in many Mediterranean archeological sites (Grindon and Davison 2013); furthermore, the limestone they dwell upon had been used for a long time as a building material and transported in large blocks along historical trading routes across the Mediterranean Sea (Fiorentino et al. 2008). The geographic distributions of haplotypes 27 and 30 (Rhodes, Israel and continental Turkey) and that of the closely related haplotypes 17 (Karpathos) and 29 (Israel) are particular striking and advocates for bringing the Knights Hospitaller of St. John back into the play. After the rising power of Islam expelled in 1291 the Knights from Jerusalem (where haplotypes 27 and 29 are found), they conquered Rhodes (where we found haplotype 27), the neighboring islands (i.e. Karpathos with haplotype 17) and the coast of Anatolia (haplotype 30). The Knights kept the Dodecanese and the Anatolian ports facing it under their control for 200 years before being defeated in 1522 by Sultan Suleiman the Magnificent and forced to withdraw to Malta (Mayer 2005;Murray 2006). During this period of time, they built anew or fortified with huge walls the already existing castles. In particular, they fortified the city of Rhodes with the Palace of the Grand Master where one of the two only umbilicate (spiriplana) populations of the island are found (Glaubrecht 1993a(Glaubrecht , 1993b; haplotype 27 (co-occurring on the island and Israel) is indeed carried by land snails collected on the rampart of one of the city gates (D'Amboise gate; Table 1). It is thus not difficult to envision these historical events as being responsible for the unusual geographic distribution of some of the mtDNA haplotypes we identified in the study (Figs 1 and 2). It would be interesting to expand this study to include samples of Levantina from Cyprus; the island was never connected to the Anatolian mainland but served as the first stronghold or retreat of the Crusaders after they had to leave the Levant following the fall of Acre in 1291.
We are aware that the scenario presented in here -although fascinating -is not the only possible one. Due to the sub-optimal quality of most of the samples at our disposal, we were able to sequence short gene fragments. This implies that we could have easily missed out on rare genetic variants. In addition, Örstan (2004) already suggested that three disjunctive records of L. spiriplana in western Turkey, just north of the region considered of this present study, could be due to introduction by humans during the Ionian period, perhaps on ballast rocks. Similarly, Welter-Schultes (1998) suggested that some Albinaria species that are found aestivating on rocks in Crete might have been carried to places outside their natural ranges on rocks used for construction or as ballast in ships. More recently, the same author (Welter-Schultes 2008) provided direct evidence that land snails have been carried on ships for more than 3,000 years in the Mediterranean area by describing shells discovered in the underwater archaeological excavations of a Late Bronze Age (3,300 years BP) shipwreck at a Southern Turkey location. Däumer et al (2012 and references therein) revealed a complex scenario in the invasive land snail Theba pisana pisana suggesting that primarily human activities rather than natural processes have shaped (and still are) the distribution of the taxon. The authors also suggested that different lineages identified on genetic bases only could have different adaptive and invasive potentials, unveiling a complex scenario where different forces at different levels (from the ecological to the genomic one) could come into play.
The data presented in here, along with the similar evidence existing for the area mentioned in the previous paragraph, suggest that two different layers of complexity (natural colonization vs. historical human activities) should be considered when addressing puzzling distributions in an area interested by intense human activi-ties since historical times. Also, this study represents a starting point for further investigations based on a more extensive sampling in terms of geographic and taxon coverage as well as molecular markers.