Mollusc species from the Pontocaspian region – an expert opinion list

Abstract Defining and recording the loss of species diversity is a daunting task, especially if identities of species under threat are not fully resolved. An example is the Pontocaspian biota. The mostly endemic invertebrate faunas that evolved in the Black Sea – Caspian Sea – Aral Sea region and live under variable salinity conditions are undergoing strong change, yet within several groups species boundaries are not well established. Collection efforts in the past decade have failed to produce living material of various species groups whose taxonomic status is unclear. This lack of data precludes an integrated taxonomic assessment to clarify species identities and estimate species richness of Pontocaspian biota combining morphological, ecological, genetic, and distribution data. In this paper, we present an expert-working list of Pontocaspian and invasive mollusc species associated to Pontocaspian habitats. This list is based on published and unpublished data on morphology, ecology, anatomy, and molecular biology. It allows us to (1) document Pontocaspian mollusc species, (2) make species richness estimates, and (3) identify and discuss taxonomic uncertainties. The endemic Pontocaspian mollusc species richness is estimated between 55 and 99 species, but there are several groups that may harbour cryptic species. Even though the conservation status of most of the species is not assessed or data deficient, our observations point to deterioration for many of the Pontocaspian species.


Introduction
The aquatic Pontocaspian (or Ponto-Caspian) biota is constituted by taxa that evolved in saline water bodies in the Caspian Sea -Black Sea -Aral Sea region and surrounding rivers in the past few million years. They include diverse groups such as diatoms, dinoflagellates, foraminiferans, crustaceans, molluscs, as well as fish and the Caspian seal. Major Pontocaspian habitats are located in the northern coastal zone of the Black Sea (mostly confined to the Romanian and Ukrainian coasts) and the Sea of Azov (mostly in the Taganrog Bay), cover the entire Caspian Sea and, until recently, the Aral Sea ( Fig.  1). However, Pontocaspian habitats are impacted by human activities such as pollution, habitat modification and introduction of invasive species (Bologa et al. 1995, Zolotarev 1996, Zaitsev and Mamaev 1997, Gomoiu et al. 2002, Grigorovich et al. 2003, Occhipinti-Ambrogi and Savini 2003, Barannik et al. 2004, Shalovenkov 2005, UNEP 2006, Stolberg et al. 2006, Selifonova 2008a, b, Popa et al. 2009, as well as the entire obliteration of environments in the case of the Aral Sea in the second half of the 20 th century (Mainguet and Létolle 1997, Andreeva and Andreev 2003, Plotnikov et al. 2016. Faunas in the Pontocaspian region have strongly changed in the past century. Pontocaspian species that were abundant only a century ago, such as Dreissena elata and D. caspia in the Caspian Sea, have vanished in the mid-20 th century (Kosarev and Yablonskaya 1994). For the Aral Sea, the faunas appear to have largely disappeared with the demise of the lake system since the 1950s (Andreeva and Andreev 2003). Abundances of several other species in the Caspian Sea and Black Sea Basin have severely declined (Bologa et al. 1995, Zaitsev and Mamaev 1997, Barannik et al. 2004, Popa et al. 2012. However, we cannot evaluate the extent or nature of biodiversity loss as there is no general agreement on the species that it might concern. Much of the diversity in Pontocaspian mollusc groups is contained within a limited number of genera. Changing taxonomic approaches through time (e.g., Zhadin 1952, Alexenko and Starobogatov 1987, Sitnikova and Starobogatov 1999, Munasypova-Motyash 2006a, b, Anistratenko 2007b, Kijashko in Bogutskaya et al. 2013, Vinarski and  combined with large morphological variability and few diagnostic characters in certain groups, as well as the paucity of living material and partial disappearance of type material, has precluded critical reassessment of species boundaries and thus species richness. For the Caspian Sea, multiple efforts to collect fresh material in the past decade failed to produce sufficient living material to elucidate these taxonomic matters for most of the groups. Sampling efforts include coastal sampling around Turali, Russia (FW, 2003); northern Azerbaijan (FW, 2016), middle and southern Azerbaijan (VA, ML, AFS, TW, 2017); Mangyshlak region coastal areas, Kazakhstan (OA, VA, 2016, 2017; the transition of the northern to middle Caspian Sea Basin in Kazakhstan (PRIDE expedition, 2017); and the Gorgan Bay in Iran (AFS, 2018). A faunal inventory of the deep-water southern Caspian Sea Basin (> 200 m water depth) of southern Azerbaijan was published lately by Mirzoev and Alekperov (2017). We are uncertain whether it concerns living material nor can we assess the species identities. Their records are mentioned below but require further confirmation. We did find some living endemic species ourselves, and from coastal areas low numbers of such species have been reported elsewhere (e.g., Latypov 2015). Yet, many species and even groups of species (e.g., Turricaspia species) have not been encountered alive despite our attempts. Our inability to collect life specimens for several groups has made a combined molecular-morphological approach to delineate species impossible. As a result, a reliable estimate of the number of species involved is lacking, and therefore the potential magnitude of the biodiversity decline is speculative. Hence, we need an alternative approach to outline the species boundaries and estimate the numbers affected.
By pooling all insights, data (published and unpublished) and expert opinions on the Pontocaspian mollusc species through taxonomists we aim to provide a list of Pontocaspian mollusc species that can serve as a base for further research. We use molluscs as a model group since they are (1) an important, representative and well-known part of the Pontocaspian fauna, (2) have a number of taxonomic specialists available, and (3) can often be identified based on their shell characters even when living populations have vanished. The Pontocaspian aquatic mollusc species list will highlight uncertainties in species complexes as to give guidance to further research in resolving taxonomic matters. The aim of this work is to compile a list of Pontocaspian mollusc species with the underlying arguments why we consider these species as (likely) valid species, to outline taxonomic uncertainties and to provide an updated estimate of species richness.

Materials and methods
A preliminary Pontocaspian mollusc species list was assembled during a PRIDE program workshop in Giessen, Germany, in May 2018. The PRIDE ("Drivers of Pontocaspian Rise and biodiversity Demise") program is an EU funded Innovative Training Network that studies the drivers of the rise and demise of Pontocaspian faunas. Using listings in Vinarski and Kantor (2016) supplemented with further information from the participants, this initial list was then circulated among a wider community of taxonomic workers for further updates and comments. Data on distribution and type material were derived from Vinarski and Kantor (2016) and further completed and amended. The systematic order above the species level follows Bouchet et al. (2017) and MolluscaBase (2018a). In cases where we deviate from the supraspecific classification, arguments are discussed below.
The list comprises aquatic Holocene Pontocaspian mollusc faunas. A substantial number of Pontocaspian species has been described from empty shells from beach material or derive from grab samples. Such samples typically are dominated by time-averaged Holocene shell assemblages, which may or may not yield living specimens and in very rare occasions also contain older (Pleistocene) material (see, e.g., Leroy et al. 2018). For the Black Sea Basin, the Holocene time interval largely coincides with the date of the marine flooding through the Bosphorus and subsequent marginalisation of Pontocaspian species to the NW coastal zone (Danube Delta to Dnieper Estuary) and the Sea of Azov (Mordukhay-Boltovskoy 1960). For the Caspian Sea, the time interval corresponds to the so-called Novocaspian period that started after the very deep Mangyshlak regression 8 ka (Fedorov 1953, Nevesskaja 1958, Yanina 2005. The time interval contains the earliest impact of humans on native faunas, such as the introduction of Cerastoderma glaucum in the Caspian Basin during the early Holocene (Fedorov 1957, Yanina 2009. It also contains the large faunal changes of the 20 th century related to pollution, invasive species, and obliteration of habitats (Kosarev and Yablonskaya 1994).
One of the greatest difficulties is to establish the identities of taxa reported as geographic subspecies. Many species have forms, varieties, and subspecies described from the Aral Sea, the Caspian Sea Basin, and the Black Sea Basin (including the Azov Sea). Often, such distinctions are made based on the geographical isolation alone or on a range of morphological characters whose variation seems to be overlapping in geographical subpopulations. In order to assess whether the geographical populations are indeed species, we need combined morphological, ecological, and molecular data, but only few studies produced this information to date (e.g., Popa et al. 2012 for Black Sea Basin Monodacna). For the Aral Sea, we expect difficulties to obtain fresh material of almost all species for molecular analyses due to the obliteration of most of the lake and its fauna in the 20 th century (Andreeva and Andreev 2003, Plotnikov et al. 2016. To date, hardly any molecular data on closely related species that are (potentially) shared between the Caspian and Black Sea have been published with the exceptions of Dreissena grimmi/D. bugensis (e.g., Therriault et al. 2004, Stepien et al. 2013 and Ecrobia maritima/E. grimmi (Haase et al. 2010). For several potentially shared species, ecological tolerances and preferences between Caspian and Black Sea Basin populations are overlapping, but in some cases (like for D. grimmi/D. bugensis) they are not. We have adopted a conservative approach, and as long as no additional arguments (morphological, ecological, or genetic differences) were found, we consider the Aral, Caspian and Black Sea varieties/subspecies synonyms. Another difficulty in especially Caspian taxa is the erection of so-called "bathymetric" subspecies, which seem to be distinguished mostly based on their depths of occurrence. As long as no other (morphological, genetic) arguments are available, we synonymise such bathymetrical forms.
A listing of synonyms and important past misidentifications from the literature is given. The list is not exhaustive and intended to show major shifts in taxonomic thinking about Pontocaspian and invasive species. The format of synonymy lists follows mostly suggestions of Matthews (1973). Asterisks in front of a record indicate valid first descriptions, a superscript "o" a prior yet invalidly introduced synonym. The status of each species is defined according to criteria outlined in Table 1.
As for the conservation status we have indicated the IUCN Red List status as of July 2018 from [www.iucnredlist.org] and added our own observations. For updated stratigraphic terminology and age estimates we refer to Krijgsman et al. (2019).

Pontocaspian
Centre of evolutionary history in Pontocaspian lakes Native Present in the Pontocaspian region today and in the Quaternary (not introduced by man) but centre of evolution not necessarily in that region: e.g., planorbid species with a Palearctic distribution, Cerastoderma glaucum.

Introduced
Species introduced in the Pontocaspian from elsewhere, usually anthropogenic: some Pontocaspian species have migrated between Pontocaspian basins and their status is explained in detail there (e.g., Monodacna colorata/Dreissena bugensis: introduced in Caspian from natural ranges in Black Sea Basin). Invasive Species that have become disruptive in the ecosystem after introduction.
Distribution. Danube Delta region and NW Black Sea Basin coastal areas of Ukraine. Taxonomic notes. We are uncertain about the status of Adacna fragilis Milaschewitch, 1908. The Black Sea Basin material has a wide variety of shapes and often is thinner and sometimes more elliptical than the Caspian A. laeviuscula. Both forms were synonymised by Graf and Cummings (2018) and indicated as a possible synonym in MolluscaBase (2018b). However, the Black Sea Basin occurrences are recorded from (coastal) lakes and small rivers suggesting little or only partial overlap in the ecological (and especially salinity) preferences of A. laeviuscula (e.g., Munasypova-Motyash 2006a, b, Popa et al. 2009. We are uncertain if A. fragilis might constitute a geographical subspecies (a status advocated by Grossu 1973), and further molecular analyses are needed to clarify the status of the Black Sea taxon.

Status.
Pontocaspian species, endemic to Caspian Sea and Aral Sea; likely disappeared from the latter. Type locality. The northern Caspian Sea and the Aral Sea . Distribution. Aral Sea (probably extinct there; Andreeva and Andreev 2003), Caspian Sea.
Taxonomic notes. Graf and Cummings (2018) consider this species as a synonym of A. vitrea, but Kijashko in Bogutskaya et al. (2013) regards it as a valid species. The latter considers A. minima minima from the Aral Sea and A. minima ostroumovi syn. n. from the Caspian Sea as distinct geographical subspecies. The likely disappear-ance of the species from the Aral Sea makes a molecular assessment of their distinctness very difficult and given the lack of other arguments we synonymise both. Furthermore, we are uncertain about the status of the subspecies Hypanis minima sidorovi Starobogatov, 1974 from the western Aral Sea. Without further data we assume it concerns a form that falls within the wide morphological variation of A. minima. We moreover are very uncertain as to the status of Hypanis minima from Holocene deposits of Aral Sea as illustrated by Filippov and Riedel (2009, fig. 4s, t). The juvenile specimen has relatively strong cardinal teeth, onset of clear ribs, and a general shape that more resembles Monodacna caspia.
Remarks. The species has been recorded mostly from the middle and southern Caspian Sea and more rarely from the eastern areas in the northern Caspian Sea down to 35 m water depth  as well as from the Aral Sea from where it may have disappeared.
Conservation status. Not assessed.

Adacna vitrea (Eichwald, 1829)
Distribution. Black Sea Basin (also in Azov Sea and adjacent lower Don River), Caspian Sea Basin, and Aral Sea (including delta of Amu-Darya River). The Aral populations may have gone extinct in the 1980s (Andreeva and Andreev 2003).
Taxonomic notes. The species has been subdivided into three geographical subspecies which were not recognised by Graf and Cummings (2018). It concerns a species with thin shells that yield very few diagnostic characters that show overlap. Here, we synonymise the subspecies pending molecular assessments of their status.

Cerastoderma glaucum
Taxonomic notes. DNA studies have shown a strong structuring between Atlantic-western Mediterranean, Ionian, and Aegean-Pontocaspian populations of C. glaucum (Nikula andVäinölä 2003, Sromek et al. 2016). According to Sromek et al. (2016: 515), the "strong genetic differentiation and the occurrence of private alleles may hint at the presence of cryptic species within C. glaucum". For a discussion on the authority of C. glaucum, see Vinarski and Kantor (2016: 69-70).
Remarks. The arrival of Cerastoderma glaucum in the Caspian Sea circa 8000 years ago has been linked to human settlement expansion through the Manych corridor (Fedorov 1957, Yanina 2009 Taxonomic notes. This concerns a common rhomboid-shaped species in the Pontocaspian region whose name is uncertain. It has a short ligament in common with C. glaucum and the persistent occurrence of ribs on the posterior margin, the well-defined character of the ribs and the regular occurrence of scales in common with western European C. edule. This form has been often referred to as C. rhomboides (Lamarck, 1819) that has been described from the Italian Pliocene but that concerns a typical glaucum form (Fig. 2), not the rhomboid form of the Pontocaspian Cerastoderma. The species has been named Cardium edule var. nuciformis by Milaschewitch (1916), but that name is a junior primary homonym of Cardium nuciforme d'Orbigny, 1850. Even though some morphological features mentioned in the description of C. lamarcki (Reeve, 1845) may resemble those of the Pontocaspian species, the former has been traced to southern Great Britain from where molecular analyses only show the presence of C. glaucum (Nikula and Väinölä 2003).
Conservation status. Not assessed.  Logvinenko and Starobogatov (1969) reported Didacna baeri from the southern basin (mostly on the eastern side) and from the middle basin down to 60 m water depth.
Taxonomic notes. In recent works (e.g., Kijashko in Bogutskaya et al. 2013), the species Didacna crassa   [= D. eichwaldi (Krynicki, 1837)] has been considered a synonym of D. baeri. However, both species can be distinguished. Didacna baeri has a less extended, more roundish shell, a less developed keel, and a low top with less projecting beak and in general more ribs than D. eichwaldi (Fig. 3). Didacna baeri occurred for the first time in the Novocaspian transgressive deposits whereas D. crassa already occurred in the late Khvalynian (Late Pleistocene). Both became very common during the Novocaspian.
Conservation status. Not assessed.
Type locality. Caspian Sea, station 116, 44°17'N, 50°22'E. Distribution. Southern, middle, and southern part of the northern Caspian Sea down to 40 m water depth, preferentially on sandy sediments .

Status.
Pontocaspian species, endemic to Caspian Sea. Type locality. "Caspium mare" (Caspian Sea) (for C. crassum . Distribution. Caspian Sea. Didacna eichwaldi is known from the middle and southern Caspian Sea basins down to 35 m water depth and cannot tolerate lowered salinities. Taxonomic notes. In recent works (Kijashko in Bogutskaya et al. 2013), the species Didacna crassa   [= D. eichwaldi (Krynicki, 1837)] has been considered a synonym of D. baeri. However, we see morphological discontinuities in our extensive material from the northern Caspian Sea Basin that implies that D. eichwaldi with its protruding umbo and shouldered appearance is distinct from D. baeri that is characterised by a rounded umbo (see discussion above under D. baeri). Despite being in common usage, the name Didacna crassa is invalid as it is a junior homonym of Cardium crassum Gmelin, 1791;Krynicki (1837) introduced Cardium eichwaldi as replacement name.
Conservation status. Not assessed.

Status.
Pontocaspian species, endemic to Caspian Sea. Possibly extinct. Type locality. Apsheron Peninsula, Azerbaijan, Quaternary. Distribution. Caspian Sea. Logvinenko and Starobogatov (1969) reported the species from the southern Caspian Sea Basin and the southern part of the middle Caspian Sea Basin down to 60 m water depth. The species has been collected from Holocene deposits and beach occurrences the western part of the middle Caspian Sea Basin as well (FW, pers. obs.). The species is reportedly extinct, not mentioned in Kijashko in Bogutskaya et al. (2013).
Remarks. The first appearance of Didacna praetrigonoides is in lower Khvalynian deposits, it became widespread during the late Khvalynian and was rare during the Novocaspian.
Conservation status. Not assessed. Didacna praetrigonoides has been reported to occur 'rarely in the modern Caspian Sea' (Nevesskaja 2007: 927), but material from recent assemblages has not been found.

Distribution.
Middle and southern Caspian Sea basins; it is most common in the middle basin at 25-85 m water depth .
Taxonomic notes. According to Logvinenko and Starobogatov (1969), two subspecies occur in the Caspian Sea at different depth ranges: D. protracta protracta at 25-50 m and D. protracta submedia  at 50-85 m. The latter differs from D. p. protracta by the relative posterior location of the umbo that is furthermore subdued. Both forms of Didacna protracta are widespread in the Khvalynian deposits of the Caspian Sea and Manych depression. According to Kijashko in Bogutskaya et al. (2013) morphological differences characteristic for the subspecies of Didacna protracta are due to allometric growth. The mere difference in depth distribution, with overlapping depths and intermediate forms, does not provide any argument to maintain these subspecies. Didacna protracta is the type species of the subgenus Protodidacna  Remarks. The authorship attribution of this species to Eichwald (1829) as proposed by several authors was rejected in Vinarski and Kantor (2016). According to them, Cardium protractum Eichwald, 1829, described from the western Ukraine, probably refers to a different species.

Status.
Pontocaspian species, endemic to Caspian Sea. Type locality. "Northern Caspian Sea on the central part of the slope" (Vinarski and Kantor 2016: 66), which likely refers to northern slope of the middle Caspian Sea Basin. 2013 Adacna (Monodacna) caspia caspia   (sub-) species that all share the relatively convex and rounded shell and well-defined ribbing. These species have been described from different areas and habitats in the Caspian Sea and have been morphologically characterised by Kijashko in Bogutskaya et al. (2013). However, neither morphological analyses of intermediate populations nor genetic analyses have been performed to clarify if the three taxa are distinct or ecomorphs of a single species. We are therefore uncertain whether M. filatovae and M. knipowitschi should be maintained.
Conservation status. Not assessed.

Status.
Pontocaspian species, native to Black Sea Basin (including lower Danube River), invasive in Caspian Sea and Volga River.
Type locality. "Hypanin fluvium, ad nigrum usque mare" [Lower course of the Yuzhnyi Bug River, all the way to the Black Sea, Ukraine].
Distribution. Native to all Black Sea Basin Pontocaspian habitats and lower courses of adjacent rivers such as the Danube, Dnieper, and Dniester; invasive in Caspian Sea Basin and lower Volga, as well as Lake Balkhash (Kazakhstan). Occurs hundreds of kilometres upstream in major tributaries (Danube: Popa et al. 2009; recent observations in Volga River upstream Volgograd by MV and AFS).
Taxonomic notes. Monodacna colorata appears to be a morphologically very variable species. Here, we propose to synonymise several local Black Sea species with this taxon. Given the difficulty to distinguish relatively flat shells typically associated with M. colorata from the more convex shells typically associated with M. pontica in, e.g., Lake Razim (Romania) and the apparent lack of genetic differentiation of convex specimens from M. colorata we assume that M. pontica is a synonym of M. colorata. Shell differences have been attributed to substrate differences. Further investigations to confirm the synonymy are required. Monodacna angusticostata was synonymised by Popa et al. (2012) based on molecular evidence, even though some morphological distinction was reported from M. colorata, which they attributed to differential habitat preference (sediment type).
Conservation status. Not assessed.
Taxonomic notes. This species has been reported in much of the 20 th century literature as Abra ovata (Philippi, 1836), which is invalid since the original name (Erycina ovata Philippi, 1836) represents a junior primary homonym of Erycina ovata Gray, 1825. Remarks. The first transfer of Abra segmentum into the Caspian Sea occurred in 1947-1948, and the species has not been detected since 1955 (Latypov, 2015).
Conservation status. Not assessed.

Status. Native/Invasive Pontocaspian species.
Type locality. Euphrates River. Distribution. Native to large parts of western Asia (including southern Caspian river systems) and northern Africa, introduced in 1939 to southern North America and in 1980 from there to Europe (Seddon and Van Damme 2016). The species has been recently recorded from the Caspian Dagestan coast (Nabozhenko and Nabozhenko 2016).
Remarks. This species has been native to south Caspian rivers including the Kura river system (Zhadin 1952) and has expanded several times in the Late Pleistocene into the Caspian Sea, where in time intervals it survived in proximal lacustrine habitats. A recent introduction and expansion of the species has been recorded in the Kizlyarsky Gulf in Dagestan (Nabozhenko and Nabozhenko 2016) and the strong increase in fresh material found around the gulf in subsequent years, including whole specimens (AS Gasanova, Makhachkala, pers. comm.) suggests the species may have established there.

Family Dreissenidae Gray, 1840
Remarks. Pontocaspian dreissenid taxonomy suffers from a lack of coordinated shell and DNA analyses. A large part of our considerations relies on the work of Rosenberg & Ludyanskiy (1994) who examined and illustrated all type material of Pontocaspian Dreissena.
Distribution. Endemic to western Ukrainian liman coast, introduced in Danube Delta, Azov Sea, Volga catchment, western and central Europe, and North America (Orlova et al. 2005, Coughlan et al. 2017. Taxonomic notes. This species has been considered as a subspecies of D. rostriformis (Deshayes, 1838) by some authors (e.g., Orlova et al. 2005), yet we follow the argumentation of Kijashko in Bogutskaya et al. (2013) to consider it as a distinct species. The proposed synonymy of Caspian D. rostriformis (= D. grimmi) and Black Sea D. bugensis by Stepien et al. (2013) Andrusov, 1897). However, syntypes of the latter illustrated in Rosenberg and Ludyanskiy (1994, fig. 3f ) show a broad and keeled Dreissena that has major morphological characters in common with D. polymorpha/elata rather than D. caspia. Filippov and Riedel (2009) reported Dreissena caspia from Holocene core deposits of Aral Sea, but given the juvenile status of their material they noted they were uncertain whether it might comprise D. polymorpha. Dreissena caspia was reported alive from the remaining "small Aral Sea" by Plotnikov et al. (2016). However, this latter record concerns more likely D. polymorpha and needs confirmation. Andreeva and Andreev (2003) mentioned that this subspecies has not been found in the Aral Sea since 1989.

Dreissena elata
Type locality. Kuuli Cape, Dazmyk, Apsheron Peninsula, Azerbaijan (Vinarski and . Distribution. Caspian Sea. Probably extinct. Taxonomic notes. Dreissena elata has morphological features in common with D. polymorpha, including a relatively wide shell and a well-pronounced keel located close to the ventral margin. However, the D. elata shell is in general wider, flatter, and has a more rounded abapical margin even though shell characters are higly variable. Dreissena elata has been reported from areas in the Caspian Sea with salinities well above 5 ‰, which is unusual for D. polymorpha elsewhere. We are uncertain whether D. elata might be a sibling species. Its apparently distinct morphology and autecological preferences suggest it is different from D. polymorpha, but it will require molecular comparison to investigate whether it concerns a mere morph that has undergone "ecological release" (Kohn 1972) or is a different species. However, no living specimens of D. elata have been recorded since 1957 (Kostianoy and Kosarev 2005) when its Caspian habitats were invaded by Mytilaster minimus.
Conservation status. Not assessed. It was reported as extinct by Kostianoy and Kosarev (2005, and references therein). If D. elata is accepted as a valid species, it might qualify for the same conservation status as D. caspia (critically endangered, possibly extinct; von Rintelen and Van Damme 2011b). Taxonomic notes. This Caspian species is very often cited as Dreissena rostriformis. Rosenberg andLudyanskiy (1994: 1497) discuss the uncertainties of this attribution but state that "D. pontocaspica, D. distincta, D. compressa, and D. grimmi are synonyms of D. rostriformis" even though they find "some justification for maintaining a distinction between an extinct subspecies, D. rostriformis rostriformis and a living one, for which D. rostriformis grimmi is the oldest name". Their figure of the lectotype of D. rostriformis (Rosenberg and Ludyanskiy 1994: fig. 2a), which derives from Pliocene deposits of the Black Sea Basin, concerns a relative small, thick-shelled, and low Dreissena with a pointed beak and lacking a keel. On interior view, the shell area outside the pallial line is thick. Deshayes's lectotype has several characters in common with modern Caspian D. rostriformis and the closely related Black Sea Basin D. bugensis. Yet, the Pliocene form has a broader umbonal area that results in a more subquadrangular shape, which is different from the modern Caspian Dreissena that have tear-drop to pear-shaped shells. The subquadrangular shape of Deshayes's material is even more pronounced in the pallial line on the shell's interior, a feature not seen in any modern Caspian material. The Pliocene Black Sea D. rostriformis has its general shape in common with Apsheronian (Early Pleistocene) Caspian dreissenids referred to as D. carinatocurvata as illustrated in Kolesnikov (1950, pl. 14, figs 14-16). Hence, we conclude that the recent Caspian species should be treated different from Pliocene D. rostriformis and the name D. grimmi should be applied instead.

Dreissena grimmi (Andrusov, 1890)
Various subspecies have been attributed to Caspian Dreissena rostriformis (see, e.g., Kijashko in Bogutskaya et al. 2013 for a synonymy list). Even though morphological differences appear to be large, intermediates are known between the morphs. Stepien et al. (2013) reviewed molecular evidence for species boundaries within Dreissena. They concluded that (1) all Caspian Sea forms that have been mentioned in literature as (sub-) species of D. rostriformis (= D. grimmi) are one and the same species and (2) there is not enough molecular evidence and great difficulty in morphology to separate the Caspian species from the Black Sea Basin D. bugensis. We agree with the first point made by Stepien et al. (2013); all forms reported from the middle and southern Caspian Sea basins appear to be mere morphs of a single species, a feature also noted by Rosenberg and Ludyanskiy (1994). However, we disagree with their second proposal. Dreissena bugensis and D. grimmi have non-overlapping ecological tolerances and are separated geographically (Rosenberg and Ludyanskiy 1994). This fact together with the very limited but consistent genetic differentiation suggests that it may concern very recently evolved sister species. In the early 1980s, D. bugensis was introduced in the Volga (Zhulidov et al. 2005) and since then spread from there to central and western Europe and North America. So far, Dreissena bugensis has only been reported from the Volga itself and its delta but not from the northern Caspian Sea Basin. If it would be conspecific with the middle-southern Caspian species, which lives at higher salinities and deeper habitats, we would expect that the invasive populations in the north would have been blended with the Caspian population in the south. With no such intermediate populations found so far we consider both taxa as viable species.

Status. Native Pontocaspian species.
Type locality. Volga and Yaik (Ural) rivers, Caspian Sea. Distribution. Eurasian (native and invasive), North America (invasive) rivers, lakes, estuaries, deltas (Rosenberg and Ludyanskiy 1994, Cummings and Graf 2015, Coughlan et al. 2017. Several unique forms/species within this group reported from the Pontocaspian region. Taxonomic notes. Dreissena polymorpha has been subject of intense DNA and ecological studies, but rarely were Caspian communities involved. Combined insights into the shell morphology, ecology, and molecular biology has to date not fully resolved several aspects of Pontocaspian records of this species. Occurrences in rivers and deltas of the Pontocaspian region are consistently attributed to Dreissena polymorpha. However, slightly deviating morphs exist(ed) in salinities typically not favoured by D. polymorpha elsewhere in the Caspian and Aral seas. A particular form of Dreissena polymorpha, documented by Kijashko in Bogutskaya et al. (2013), viz. D. polymorpha andrusovi (his figure 118b) will need further study as it has many morphological similarities with D. caspia (including general shape, location of semidiameter, and broad flat shape of hinge platform).
Conservation status. Least Concern (Van Damme 2014). ( Remarks. The species, native to the southern coast of North America, was first introduced in Europe in 1835 (Heiler et al. 2010). In the Pontocaspian region, it first appeared in the northern Black Sea Basin in 2002 and was first collected in the Caspian Sea in 2009 (Heiler et al. 2010). It is easily distinguished from Pontocaspian dreissenids by the presence of an aphophysis near the hinge.

Gastropoda
Taxonomic notes. The latest phylogenetic data supports a sister relationship between Theodoxus danubialis and the clade containing T. fluviatilis and T. velox (AFS, unpublished data). Some authors believe T. danubialis and T. prevostianus may represent different species given some level of genetic, ecological, and morphological differentiation (Fehér et al. 2009, Welter-Schultes 2012 but see also Bandel 2001). More recent unpublished results may suggest that the genetic level of differentiation between these species is more indicative of intraspecific diversity within a single species (AFS, unpublished data).
Distribution. Widely distributed all over Europe, Anatolia, and north-western Africa. Within the Pontocaspian region, it is a common component of the lower reaches of Black and Azov Sea drainages (specifically in Bulgaria, Romania, and Ukraine). Towards the east, the species extends at least as far as the Don River system in Russia and the coastal rivers of Georgia, but it is absent from the Caspian system. Records of this species from Iran and western Asia are likely misidentifications (AFS, unpublished data).
Taxonomic notes. Theodoxus fluviatilis exhibits considerable variation in shell colouration and shape (Glöer and Pešić 2015). Unpublished molecular data confirm the synonymy of a number of taxa such as Theodoxus euxinus syn. n., T. danasteri, and T. subthermalis syn. n., and further suggest the inclusion of T. saulcyi and T. heldreichi (AFS, unpublished data). A final decision concerning the status of T. dniestroviensis Put', 1972 described from the Dniester River (Rukhotyn village, Khotyn district, Chernivtsi region, Ukraine) is not possible at the moment. Despite appropriate efforts, we were unable to trace the type specimens of this species. Based on the original description and illustration (Put' 1972) it was considered as a junior synonym of T. fluviatilis by Anistratenko et al. (1999) having an unusual colour pattern. Theodoxus milachevichi was described as a subfossil from the Crimean coast. It closely resembles morphotypes of both T. fluviatilis and T. velox V. Anistratenko in O. Anistratenko et al., 1999 and might be synonym of either species (compare type material illustrated in Kantor and Sysoev 2006). However, the morphological variability of the taxa involved, as well as the lacking possibility of acquiring genetic data for T. milachevichi, complicates a decision on the independence or synonymy of this species.
Distribution. Present along the Caspian Sea shores, in the Volga River, and the Sea of Azov. Lived until the late 1980s in the Aral Sea but is possibly extinct there now (Andreev et al. 1992, Aladin et al. 1998, Micklin et al. 2014. Taxonomic notes.  introduced the species Neritina liturata based on material from the shores of Derbent (Dagestan, Russia, northwestern Caspian Sea). That name is invalid as it is a junior primary homonym of N. liturata Schultze, 1826; it was replaced by Lindholm (1924) with Theodoxus pallasi (see also Anistratenko et al. 2017). Theodoxus pallasi is a widely used name, but a major nomenclatural change might be due. Unpublished molecular data suggest that all Theodoxus from the Caspian Sea, Azov Sea, and Armenian lakes Sevan and Yerevan, as well as several mineral springs and streams in the Khorasan provinces of Iran, belong to a single species (AFS, unpublished results). The oldest name available for that group is Theodoxus major Issel, 1865, described from Lake Sevan in Armenia (originally as variety of the unavailable name T. schirazensis). Akramovskiy (1976) noted the similarity of T. pallasi and T. major and considered the latter as a morphotype of the former. Although he did not explicitly state it, he thereby suggested the two taxa to be synonymous. This view was adopted by Vinarski and Kantor (2016), who listed major in synonymy of pallasi, although Issel's (1865) name has priority. The potential synonymy also involves T. schultzii. Despite the characteristic appearance of the syntypes, the presence of intermediate morphologies in samples taken on shores of Azerbaijan and Kazakhstan in 2016 and 2017 (pers. obs. OA, VA, FW) indicates a close relationship with T. pallasi. The radulae of these two species differ in the relative width of the central and marginal teeth (see Zettler 2007 andcompare Anistratenko et al. 2017).
Unfortunately, the types of T. major, supposed to be in the Museo Regionale di Scienze Naturali, Torino, are inaccessible at the moment due to museum renovation (E Gavetti, pers. comm., Oct 2018). We refrain from a final conclusion on the synonymy of the species involved until information on the types of all taxa as well as published molecular data are available. For details on the taxonomic relationship between T. pallasi and T. astrachanicus, see discussion in Anistratenko et al. (2017).
Distribution. Middle and southern Caspian Sea basins, between 15 and 100 m .
Taxonomic notes. See discussion of T. pallasi for notes on the potential synonymy with T. major Issel, 1865. The status of T. jukovi still requires confirmation .
Conservation status. Not assessed. Status. Pontocaspian species, name uncertain. Type locality. Dnieper Delta, Zbur'ivka liman, Ukraine. Distribution. This species was believed to be restricted to drainage systems of the northern Black Sea coast (even though the Oskol River lies far from the Black Sea coast), but unpublished molecular data suggest it may be distributed as far north as the eastern part of the Baltic Sea and as far south as Anatolia (AFS, unpublished data).

Theodoxus velox
Taxonomic notes. The species was listed as junior synonym of T. fluviatilis by . Theodoxus velox is indeed challenging to differentiate from some regional morphotypes of that species given the overlap in shell patterns. Unpublished molecular data indicate however that T. velox belongs to a different molecular clade (AFS, unpublished data). The distribution range of that clade overlaps with the range of T. sarmaticus (Lindholm, 1901), which is widely accepted as a junior synonym of T. fluviatilis in the literature (e.g., Vinarski and Kantor 2016). A revision of the taxa involved and study of the type material is required to solve the synonymy issues.
Conservation status. Not assessed.
Taxonomic notes. We find the attribution of this species to the genus Eupaludestrina unsatisfactory, yet a further revision is required to establish and stabilise the generic attribution as there is considerable confusion. It is commonly classified in the South American genus Heleobia (e.g., Prié 2011), whereas Kroll et al. (2012) suggested that this species belongs to the genus Semisalsa, a group of European Cochliopidae distinct from Heleobia. However, Semisalsa is currently listed as junior synonym of Eupaludestrina Mabille, 1877 (type species: Hydrobia macei Paladilhe, 1867, by subsequent designation by Kadolsky 2008). Following Kadolsky (2008), Eupaludestrina is currently ranked as subgenus of Heleobia in MolluscaBase (2018), but both the phylogenetic and geographic distinction of the European and American species suggest separation on the genus level.
Remarks. It is unclear whether the species is native to the Pontocaspian area or a recent immigrant.

Family Hydrobiidae Stimpson, 1865
Remarks. The Hydrobiidae form the most species-rich mollusc group in the Pontocaspian region. However, in general, useful shell characters are few and highly variable (Wilke and Delicado in press). Descriptions in the past have often been very general, and illustrations of types are notably poor for several of the endemic taxa. A strong tendency of naming large numbers of species has developed throughout the 20 th century (e.g., Logvinenko and Starobogatov 1969), but for some groups where morphological and genetic analyses could be performed (e.g., Caspiohydrobia spp.) it has been demonstrated that actual species numbers were much lower than the number of species described (Haase et al. 2010). For many of the endemic species, especially in the genus Turricaspia, the apparent loss of types, combined with the lack of living material makes it impossible to assess their taxonomic status. Currently, a number of taxonomic works is in progress on the endemic Pontocaspian hydrobiid groups, and some different insights on the genus-level classifications exist. Here, we adopt a conservative approach, mostly based on Neubauer et al. (2018).

Subfamily Caspiinae Dybowski, 1913
Remarks. The distinction of the genera Caspia, Ulskia, and Clathrocaspia follows Neubauer et al. (2018). The three taxa are differentiated based on details of the protoconch and the expression of teleoconch sculpture. Caspia s. s. is characterised by a single distinct but fine spiral keel below the suture. It is usually smooth, yet within the type species some reticulate ornament can be found. Species of Clathrocaspia expose a distinctive, reticulate pattern on the teleoconch and a malleate protoconch with faint spiral threads. The aperture of Clathrocaspia often develops a distinct flat base. The discinction of the two genera is subject of current research. Ulskia also has a malleate protoconch but with more distinct spiral threads; teleoconch sculpture is occasionally present as minute elongate nodules. Taxonomic notes. The type material is stored in the von Baer collection of Caspian Sea molluscs in the Zoological Museum of Lviv University (Ukraine) and comprises more than a hundred syntypes (Anistratenko et al. 2018). The slender shell, the presence of a fine spiral keel below the suture, and the occasionally weakly reticulated surface distinguish this species from congeners. Conservation status. Not assessed.  fig. 1(9). Caspia valkanovi (Golikov & Starobogatov, 1966. pl. 41, fig. N Taxonomic notes. The identity and status of this subfossil taxon, described from phaseoline silt, are somewhat uncertain. The holotype illustrated in Kantor and Sysoev (2006) is poorly preserved and does not allow a proper assessment of its validity. The general shape and size are indicative of the genus Caspia and it looks like a variety that might even be a synonym of C. baerii. Furthermore, we are not entirely certain as to the stratigraphic age of the stratigraphic origin of this species. The phaseoline silt is a marine Holocene unit, yet it contains reworked Late Pleistocene Neoeuxinian (Pontocaspian) species (FW, pers. obs.).

Caspia baerii
Conservation status. Not assessed. Distribution. Presently endemic to the Caspian Sea. The species was also recorded from the Holocene of Danube Delta, Ukraine (Anistratenko and Prisjazhnjuk 1992).

Clathrocaspia brotzkajae (Starobogatov in
Taxonomic notes. The species differs from its congeners in the bulbous shape, with a ratio of body whorl height/shell height of approx. 3/4, as well as regarding the expanded aperture. Conservation status. Not assessed.  (Mirzoev and Alekperov 2017, who reported the species as Turricaspia gmelinii).
Taxonomic notes. The broad shell and the heavily reticulated surface distinguish this species from congeners. Pyrgula sowinskyi, from the middle and southern Caspian Sea, and P. gaillardi, from the Caspian Sea shore between Astara and Hashtpar (= Talesh), Iran, closely resemble C. gmelinii in terms of shell shape, the shape of the aperture, and the distinct reticulate teleoconch sculpture. Very likely, the two species are synonyms of C. gmelini. Since the type material of Logvinenko and Starobogatov (1969) has not been found, and the whereabouts of the material of Tadjalli-Pour (1977) is unknown, a re-examination of these species has to be postponed. Here, we suggest to treat them as nomina dubia until more information becomes available.

Status. Pontocaspian species, identity uncertain.
Type locality. Southern Caspian Sea (no details), between 40-75 m water depth. Distribution. Endemic to the Caspian Sea. Taxonomic notes. This species hardly differs from C. pallasii and might be a junior synonym. Observations on Holocene material from the southern and northern Caspian Sea shores (VA, TN, FW) suggest that the minor differences range within intraspecific variability but further studies (preferentially involving DNA) are required to solve the identity of this taxon. The classification in Clathrocaspia is based on the reticulate sculpture typical of that genus.

Status. Accepted Pontocaspian species.
Type locality. Ukraine, in the Dniester River (exact locality not specified). Distribution. Azov Sea and northern Black Sea Basin. Known from the Holocene of Danube Delta, Ukraine (Anistratenko and Prisjazhnjuk 1992).
Taxonomic notes. Clathrocaspia knipowitschii, C. makarovi, C. gmelini aluschtensis, and C. stanislavi were all described from the northern margin of the Black Sea. After detailed morphological comparison of C. knipowitschii and C. makarovi syn. n. and preliminary genetic analyses (TW, unpublished data), we conclude that both taxa should be considered synonyms. Very likely, also C. gmelini aluschtensis and C. stanislavi are synonyms of C. knipowitschii, but a final decision on that matter requires investigation of the type material.

Status. Accepted Pontocaspian species.
Type locality. Don Delta, Russia. Distribution. Known only from the type locality. Taxonomic notes. The species has distinctive shell characters: broad conical shape with a weak subsutural bulge and apically thickened peristome.
Remarks. The type material was collected by Mordukhay-Boltovskoy in 1937 and comprises two specimens, the holotype and the paratype. Three additional specimens were collected from the same region in 2006 (Anistratenko 2007a). The salinity at the type locality fluctuates between freshwater and ca. 1‰.
Conservation status. Not assessed. In the fifty years since the description of this species five specimens have been collected; this is likely evidence of its rarity. Known only from two close localities, C. logvinenkoi appears to have an extremely narrow distributional range in the Azov-Black Sea Basin, being endemic to the Taganrog province (e.g., Anistratenko 2007a). , figs 9-21.

Status. Pontocaspian species, identity uncertain.
Type locality. Bulgaria, Danube Island Vardim (43°37'N, 25°28'E). Distribution. Only known from type locality. Taxonomic notes. This species closely resembles C. knipowitschii concerning shape, size, and sculpture. According to Boeters et al. (2015), the two species differ in the degree of cover of the umbilicus, the shape of the peristome and the size and number of whorls of the protoconch. Molecular and/or more in-depth morphological and anatomical studies are required to confirm that these apparently minor differences are sufficient to separate the species.
Remarks. If the species would be confirmed, it concerns a Pontocaspian species whose distribution currently is outside prime Pontocaspian habitat, yet Boeters et al. (2015) implied they would expect that several of the Caspia records from the lower Danube and Razim Lake complex might be attributed to C. milae as well. The Razim Lake complex is Pontocaspian habitat.
Conservation status. Not assessed.  Logvinenko and Starobogatov (1969) sketch a broad and conical shell. As such, it differs from the more elongate and ovoid Ulskia ulskii . A revision is required to clarify its taxonomic status.
Conservation status. Not assessed.

Subfamily Hydrobiinae Stimpson, 1865
Remarks. In addition to the taxa discussed below, the following species of Hydrobiinae have been mentioned from the Black Sea basin (updated statuses after MolluscaBase 2018a): Hydrobia aciculina (Bourguignat, 1876), H. acuta (Draparnaud, 1805), H. euryomphala (Bourguignat, 1876), H. mabilli (Bourguignat, 1876) (Glöer and Pešić 2012); possibly also northern and central Kazakhstan and Tajikistan , however, no molecular data are known to confirm the identity of the Central Asian snails. This species was mentioned from depths between 200 and 500 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as Caspiohydrobia curta and C. gemma).
Taxonomic notes. Most of the species that have been assigned to the genus Caspiohydrobia Starobogatov, 1970, including its type species, Pyrgohydrobia eichwaldiana Golikov & Starobogatov, 1966, range within the morphological variability of E. grimmi. Previous examination of some Caspiohydrobia juvenile shells Riedel 2009, Anistratenko 2013, fig. 4A-C) as well as reproductive systems and radula did not find any criteria to support differentiation. Probably, all of the thirty Caspiohydrobia species listed by Kantor and Sysoev (2006) and Vinarski and Kantor (2016)  Taxonomic notes. Hydrobia pontieuxini, described from the Black Sea coast in Mangalia, Romania, has been considered a synonym of E. maritima based on molecular data (Kevrekidis et al. 2005). Herein, we also consider the Pseudopaludinella species introduced by Anistratenko and Prisjazhnjuk (1992) Radoman, 1977, which have been listed for the Black Sea Basin (Anistratenko 1991, Anistratenko and Prisjazhnjuk 1992, Anistratenko et al. 2011, are junior synonyms or misidentifications of this species. Conservation status. Least Concern (Van Damme 2011a).

Subfamily Pyrgulinae Brusina, 1882
Remarks. The genus concepts of Pontocaspian Pyrgulinae follow the revision of Neubauer et al. (2018). Further change is expected in several of the keeled species here

Status. Accepted Pontocaspian species.
Type locality. At the Volga River mouth near Astrakhan, and towards the Caspian Sea; also in recently lithified fossil limestone at the shores of Dagestan, Russia.
Distribution. Caspian Sea, Azov Sea, and northern Black Sea region. This species was mentioned in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as Turricaspia variabilis, T. derbentica, and T. trivialis).
Taxonomic notes. Neubauer et al. (2018) recently demonstrated the high variability of this species. Comparison of available illustrations and descriptions of the species listed in the synonymy list indicates that all of them range within this species' variability. Consequently, we consider all of them as junior synonyms of C. variabilis. A more in-depth review of the type material of the species involved is required to confirm this approach.
The status of Paludina bogensis Dubois in Küster, 1852, which was listed as a valid species of Turricaspia by Anistratenko and Stadnichenko (1995), is still unclear. That species was described from the Zapadnyi Bug River in Poland and closely resembles C. variabilis. It is, however, unlikely that a Pontocaspian species typical of oligohaline conditions occurs so far away in a pure freshwater environment. "Paludina eichwaldi Krynicki, 1837" found in the literature is a nomen nudum. Martens (1874) provided measurements and made the name available, but he listed Paludina variabilis  in synonymy, which has priority. Dybowski (1887) obviously overlooked this and considered Nematurella eichwaldi Krynicki a valid species. We follow Vinarski and Kantor (2016) and consider the species as a junior synonym of Clessiniola variabilis.

Status. Accepted Pontocaspian species.
Type locality. Southern and western parts of the Middle Caspian Sea, 36-120 m. Distribution. Middle and South Caspian Basin. This species was mentioned from depths between 200 and 400 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as Turricaspia abichi).
Taxonomic notes. The species differs from the L. cincta in its much larger size, the conical shape, the narrower subsutural band, and the larger aperture (compare Neubauer et al. 2018).

Status. Accepted Pontocaspian species.
Type locality. Gulf of Baku, Azerbaijan. Distribution. Southern Caspian Sea . Taxonomic notes. For a detailed discussion about the identity of this species and its synonym, see Neubauer et al. (2018).
Taxonomic notes. For a detailed discussion about the identity of this polymorphic species and previous misidentifications, see Neubauer et al. (2018).

Status.
Pontocaspian species, identity uncertain. Type locality. Western part of the middle Caspian Sea, 0-50 m water depth. Distribution. Type locality only. Taxonomic notes. We cannot verify the status of this species given the inadequate descriptions and illustrations and its general resemblance to other species that were described earlier.

Status. Accepted Pontocaspian species.
Type locality. Ukraine, Danube Delta, lakes Yalpug and Kugurlui. Distribution. North-western Black Sea Basin (Anistratenko and Stadnichenko 1995). Taxonomic notes. Based on the illustration of the holotype in Kantor and Sysoev (2006), we tentatively place the species in the genus Laevicaspia. A more detailed study is necessary to clarify its systematic position.
Distribution. Endemic to the Caspian Sea. This species was mentioned from depths between 200 and 400 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as Turricaspia kolesnikoviana).
Taxonomic notes. For a detailed discussion about the identity of this species, see Neubauer et al. (2018).
Conservation status. Data Deficient (Vinarski 2011k   and middle basin (personal observation based on material from Dagestan region, TAN, FW). This species was mentioned from depths between 200 and 300 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as Turricaspia kowalewskii).
Taxonomic notes. This species differs from L. kolesnikoviana in its bigger size, broader shape, and thinner peristome. Laevicaspia cincta can be distinguished based on the stouter shape and the presence of a narrow subsutural band.
Conservation status. Not assessed.
Conservation status. Not assessed.
Distribution. Limans and lower reaches of rivers Don, Dnieper, Dniester, and Southern Bug entering the northern Black Sea Basin and the Azov Sea (Taganrog Bay), as well as in coastal lakes Kotlabukh and Yalpug . The record of an undescribed subspecies of T. boltovskoji from the Caspian Sea mentioned by Anistratenko and Stadnichenko (1995) is probably based on a misidentification.
Taxonomic notes. Golikov and Starobogatov (1966) and Alexenko and Starobogatov (1987) introduced a plethora of names for morphologically similar species from the northern Black Sea Basin, partly deriving from subfossil horizons. They differ from Laevicaspia lincta slightly in the number of whorls and outline shape, but overall range within its morphological variability. Here, we consider them tentatively all junior synonyms of L. lincta. Since Starobogatov's type material is unknown, support for this approach requires collection of new material from the type localities of these taxa. Molecular data confirmed the conspecifity of L. lincta and L. milachevitchi (Wilke et al. 2007).
Distribution. Endemic to the Caspian Sea. This species was mentioned from depths between 200 and 300 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as Turricaspia marginata).
Taxonomic notes. The status of this species is uncertain. The illustrations of the type material by Vinarski et al. (2013) suggest a tentative placement in the genus Laevicaspia. It shows close similarities with L. sieversii (Clessin in Dybowski, 1887). A careful revision of the species is required to clarify its taxonomic status and systematic placement.
Conservation status. Not assessed. Taxonomic notes. This species has not been found since its first description, its identity is unclear . Judging from the description and drawing in Dybowski (1887), we suggest a systematic placement in Laevicaspia. It might be related to L. conus   Kantor and Sysoev (2006) show a small shell with four whorls, of which the latter two bear a distinct keel. The small size and the relatively large protoconch suggest that the type specimen is a juvenile shell. More specimens (including adult material) are required to shed light on this species' identity.
Distribution. Middle and southern Caspian Sea . This species was mentioned from depths between 200 and 400 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as T. laticarinata).
Taxonomic notes. The species is characterised by a massive keel near the lower suture. ?Turricaspia dimidiata is distinguished based on its more centrally placed keel. This distinction is tentative and only based on comparison of available illustrations; we are aware of the possibility that these differences might not be diagnostic. Moreover, the keel seems to become stronger with increasing water depth (Starobogatov 1968). , which differs from T. basalis only in the strength of the keels, was considered a junior synonym by Neubauer et al. (2018).

Status.
Pontocaspian species, identity uncertain. Type locality. Western part of the Caspian Sea. Distribution. Type locality only. Taxonomic notes. The description and drawing of this species provided by Logvinenko and Starobogatov (1969) do not allow an evaluation whether it is a distinct species or synonym of a previously species.

Status.
Pontocaspian species, identity uncertain. Type locality. Ukraine, in the Dnieper Delta. Distribution. Type locality only. Taxonomic notes. The status of this species is highly uncertain. The slender conical shell illustrated by Alexenko and Starobogatov (1987) suggest classification in the genus Turricaspia, which is otherwise only known from the Caspian Sea.

Status.
Pontocaspian species, identity uncertain. Type locality. Western part of the southern Caspian Sea. Distribution. Type locality only. Taxonomic notes. The species has not been found since its first description, and the whereabouts of the type material is unknown. Logvinenko and Starobogatov (1969) illustrate a small slender shell with convex whorls. It might well be a juvenile of another species.

Status.
Pontocaspian species, identity uncertain. Type locality. Middle Caspian Sea, 25-80 m. Distribution. Type locality only. Taxonomic notes. The illustrations provided by Logvinenko and Starobogatov (1969) indicate a large conical shell with nine convex whorls and a large, slightly inflated last whorl. These features are reminiscent of T. meneghiniana (Issel, 1865). However, T. concinna has not been found since its first description. The type material has been very recently detected in the collections of ZIN and awaits further study.

Status.
Pontocaspian species, identity uncertain. Type locality. Western shore of the middle Caspian Sea. Distribution. Middle and south Basin of Caspian Sea. This species was mentioned from depths between 200 and 300 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. The status of this species is highly uncertain. The illustrations of Logvinenko and Starobogatov (1969) show a slightly distorted shell with weakly convex whorls and a thin line below the suture. We are uncertain whether it might concern a growth aberration of a more common species.

Status. Accepted Pontocaspian species.
Type locality. In fossil limestone of Dagestan, Russia.

Distribution.
Middle and southern Caspian Sea . This species was mentioned from depths between 200 and 500 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. Although there is little doubt about the validity of this species, its true identity and possible synonyms are unclear.  description clearly indicates a slender shell with median keel. His type material is unfortunately unknown. The high number of keeled species complicates an evaluation what is the "true" T. dimidiata and what are synonyms. We tentatively consider Turricaspia bakuana Kolesnikov, 1947 a junior synonym of this species, based on its slender shell with median keel matching Eichwald's description as well as the prevailing concept of T. dimidiata (compare Kantor and Sysoev 2006). More data are required to support this view.

Status.
Pontocaspian species, identity uncertain. Type locality. Eastern part of the southern Caspian Sea. Distribution. South Caspian Basin. This species was mentioned from depths between 200 and 500 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. The identity of this species is unclear. Its shell resembles T. lyrata  in terms of general shape and the large, flat protoconch; it differs from that species in the large size. The type material has been very recently found in the collection of ZIN and awaits further study. Until then, we refrain from a final decision on the species' status, but we have severe doubt that Pyrgula eburnea is a distinct species.
Taxonomic notes. There is considerable confusion about the identity of this species. Dybowski (1887) described and illustrated a very slender shell with a distinct whorl profile showing a straight-sided upper half and a convex lower half. In contrast, the illustrations in Logvinenko and Starobogatov (1969) suggest a similarly slender yet distorted shell with near almost sided whorls and expanded aperture. A restudy of the type material of T. elegantula show close similarities to T. spica. It differs from that species in the more slender outline and flattened whorls.
Conservation status. Not assessed.

Status. Accepted Pontocaspian species.
Type locality. Caspian Sea (no details). Distribution. Middle Caspian Sea Basin . This species was mentioned from depths between 200 and 400 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. The nearly straight-sided, strongly attached whorls easily distinguish this species from most other Turricaspia species. Only Turricaspia grimmi (Clessin & Dybowski in Dybowski, 1887) has a similar whorl arrangement, but its shell is slightly wider and the whorls are weakly stepped and bear a thin subsutural band.

Status.
Pontocaspian species, identity uncertain. Type locality. Western part of the middle Caspian Sea, 80 m. Distribution. Middle and South Caspian Basin. This species was mentioned from depths between 200 and 400 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. The slender elongate shell with whorls slowly increasing in height distinguishes this species from its congeners. However, a proper assessment of the species' status requires investigation. The whereabouts of the type material is unknown and no other records of this species are known, so we are not able to verify the status of this species.
Conservation status. Not assessed.  . This species was mentioned from depths between 200 and 300 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. The peculiar morphology with straight-sided, weakly stepped whorls with a thin subsutural band is unique among Caspian Pyrgulinae. See above for a comparison with T. eulimellula.

Status. Accepted Pontocaspian species.
Type locality. Caspian Sea (no details). Distribution. Endemic to the Caspian Sea (after Logvinenko and Starobogatov 1969); it occurs in the western part of the middle and southern Caspian Sea basins, but these authors used a slightly different concept of the species. This species was mentioned from depths between 200 and 300 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as Turricaspia lirata).
Taxonomic notes. See Neubauer et al. (2018) for a detailed discussion of the species and its synonyms.
Conservation status. Not assessed.
Distribution. Type locality only. Taxonomic notes. The species can be distinguished based on its extremely slender shell with whorls slowly increasing in size. Still, clarification of its identity as well as its generic classification requires investigation of additional material.
Conservation status. So far only known from Holocene deposits of the type locality; species might be extinct. Within Holocene deposits in the Black Sea small amounts of reworked Late Pleistocene "Neoeuxinian" faunas are found (FW, pers. obs.), and therefore the stratigraphic origin of such Pontocaspian species is uncertain.

Turricaspia meneghiniana (Issel, 1865)
Type locality. Caspian Sea (no details). Distribution. Southern Caspian Sea . This species was mentioned from depths between 200 and 500 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. The very slender shape and the characteristic, highly convex whorls that slowly and regularly increase in height distinguish the species from most congeners. Pyrgula vinogradovi Logvinenko &Starobogatov, 1969 andP. astrachanica Pirogov, 1971, which show very similar traits, might be junior synonyms. A more indepth study is required to solve their statuses.
Conservation status. Data Deficient (Vinarski 2011l Distribution. Type locality only. Taxonomic notes. This species was originally attributed to the genus Caspia due to its small shell. A study of the holotype of this species, specifically its protoconch characteristics, suggest placement in the genus Turricaspia. Further studies are required to assure its validity. Remarks. The species is known only from the holotype. The occurrence of Turricaspia in the Black Sea Basin is unusual, as almost all other pyrguline Black Sea species are assigned to the genus Laevicaspia (but see remark at T. spica for another unusual occurrence).
Taxonomic notes. The slender shell with a keel near the lower suture is reminiscent of T. basalis . The short description and poor drawing precluded the verification of its status. The type material has been very recently detected in the collection of ZIN and awaits further study.
Conservation status. Not assessed.

Status.
Pontocaspian species, identity uncertain. Type locality. Caspian Sea (no details). Distribution. Southern Caspian Sea . This species was mentioned from depths between 200 and 300 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. The identity of this species is uncertain. Dybowski and Grochmalicki (1915) describe and illustrate a shell with eight convex whorls bearing a weak, hardly protruding, irregular shaped keel near the lower suture. According to these authors, the keel varies considerably between a thin thread, a blunt bulge, or a weak thickening at the suture. In contrast, the drawings provided by Logvinenko and Starobogatov (1969) and reproduced by Kantor and Sysoev (2006) suggest a shell with straight-sided whorls and a distinct keel. Inspection of the type material is required to clarify the status of this species.
Conservation status. Not assessed.
Conservation status. Not assessed.

Status. Accepted Pontocaspian species.
Type locality. Caspian Sea (no details). Distribution. Endemic to the Caspian Sea, reported from the middle and southern Caspian Sea basins . This species was mentioned from depths between 200 and 300 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. The species can be easily distinguished from other Turricaspia species based on its relatively broad shell, the low-convex whorls, and its small size .

Status. Accepted Pontocaspian species.
Type locality. Caspian Sea (no details). Distribution. Endemic to the Caspian Sea, reported from the western part of the middle Caspian Sea .
Taxonomic notes. The very characteristic tripartite whorl profile allows an easy identification and discrimination from other Pontocaspian Pyrgulinae .
Conservation status. Data Deficient (Vinarski 2011n fig. 360(4). 2016 Turricaspia sajenkovae  Taxonomic notes. Judging from the drawing in Logvinenko and Starobogatov (1969), presenting a small slender shell with ca. 5.5 highly convex whorls, the species might be based on a juvenile specimen. It might be a junior synonym of the similarly shaped T. meneghiniana (Issel, 1865). Without investigating the type material, which has not been found in the ZIN collection, the identity of this species remains unclear.
Conservation status. Not assessed.
Taxonomic notes. As for the previous species, it is highly uncertain whether this taxon is a distinct species. It might also be based on a juvenile and could be a synonym of an earlier described species, perhaps T. pulla or T. lyrata.

Status. Accepted Pontocaspian species.
Type locality. Western part of the middle Caspian Sea. Distribution. Middle and southern Caspian Sea. This species was mentioned from depths between 200 and 300 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017).
Taxonomic notes. The fast growing whorls terminating in a large body whorl with expanded aperture are characteristic for this species and facilitate discrimination from other Turricaspia species.
Conservation status. Data Deficient (Vinarski 2011u Taxonomic notes. As the oldest described species presently attributed to Turricaspia, the validity of this species is without doubt. Its identity, however, is poorly known, given the limited information and poor drawing provided by Eichwald (1855), as well as the largely diverging concepts applied by later authors (see Neubauer et al. 2018 for a detailed discussion of the matter). We have a geographic record (Anistratenko and Prisjazhnjuk 1992) that is outside the Caspian-Aral distribution range of this genus. Comparison of the Danube material with Caspian specimens suggests the identification might be correct, yet further detail study is required to assess whether the Danube record might actucally not be an unusual form of Laevicaspia lincta.
Taxonomic notes. The species is characterised by a slender conical shell with weakly convex whorls with weak subsutural swelling and a slightly inflated body whorl with large aperture.

Status.
Pontocaspian species, identity uncertain. Type locality. Eastern part of the northern Caspian Sea. Distribution. Type locality only. Taxonomic notes. Logvinenko and Starobogatov (1969) illustrated a comparably small shell with eight highly convex whorls, large body whorl, and large aperture. Reliable assessment of the species' status requires investigation of the type material, which has only been discovered in ZIN in June 2018 and awaits further study.
Taxonomic notes. The species was recently described in detail by Neubauer et al. (2018). The species was distinguished from A. dybowskii and A. andrusovi by the higher spire, but this is a variable character. Currently, the taxonomy of Andrusovia species is the subject of further study.
Remarks. Starobogatov (2000) mentioned that the type material is housed in the ZIN collection, but we were unable to find the holotype and it is presumed lost. Only recently, paratypes of this species were detected at the Zoological Museum of Moscow University and are currently being studied.
Distribution. Middle and southern Caspian Sea (Starobogatov 2000). This species was mentioned from depths between 200 and 400 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as Horatia marina).
Taxonomic notes. According to Neubauer et al. (2018), this species might be a senior synonym of A. brusinai Starobogatov, 2000. Inspection of recently discovered type material appears to support that view, but more in-depth studies are required to evaluate the status of this species.
Remarks. The holotype is not traced and presumed lost. Only recently, paratypes of this species were detected at the Zoological Museum of Moscow University and are currently being studied.
Conservation status. Not assessed.
in Romania. The shell they later illustrated (Golikov and Starobogatov 1972) indeed shows similarities with T. dimidiata. The shell illustrated in Grossu (1951) fig. 11a-c, pl. 3, fig. 10a  . This species was mentioned from depths between 200 and 300 m in the South Caspian Basin of Azerbaijan (Mirzoev and Alekperov 2017, who reported the species as Anisus colesnikovi [sic]).
Taxonomic notes. Logvinenko and Starobogatov (1966b) considered this species and Andrusovia dybowskii Brusina in Westerlund, 1902b to belong in the same genus, Anisus (Andrusovia), rendering P. dybowskii  a junior homonym. Therefore, they introduced A. kolesnikovi as replacement name. Since both taxa do clearly not belong to the same genus or even the same family, the replacement name is to be discarded.
The species resembles G. eichwaldi regarding the general habitus; it differs in the more pronounced angle at the transition between whorl flank and apical plane. A revision is required to investigate if the Caspian Gyraulus species are distinct species or morphotypes of G. eichwaldi. The generic placement follows Vinarski and Kantor (2016). Note that those authors listed the earlier described P. eichwaldi dybowskii    . Taxonomic notes. The species in its present combination as Gyraulus sulcatus (following Vinarski and  is invalid as it is a secondary homonym of the Miocene Gyraulus sulcatus (Hilgendorf, 1867). We refrain here from introducing a replacement name as the species' status is uncertain. It resembles G. eichwaldi and G. kolesnikovi in outline shape and differs only in the more pronounced angle between whorl flank and apical plane and the shallow furrow on the apical side. An in-depth revision is required to clarify if Gyraulus sulcatus is a distinct species or a mere morphotype of G. eichwaldi (Clessin & Dybowski in Dybowski, 1887).
Conservation status. Not assessed.

Discussion and conclusions
The annotated check-list presented here is a first attempt to assess the species diversity of the Pontocaspian molluscs by experts working in different countries and fields (neontology, palaeontology, biogeography, phylogenetics). Hitherto, progress has been limited by a number of factors: (1) fresh material for genetic studies is available only for few nominal species, and (2) the type series of many species are lost or at least have not yet been found. This concerns not only the species described by Eichwald or Grimm in the 19 th century; the type specimens of many species established by Starobogatov and his co-workers in the 1960-2000s could not be traced in ZIN (Kantor andSysoev 2006, Vinarski and. Furthermore, progress has been limited by (3) a lack of representative shell samples to undertake quantitative statistical analyses of conchological variation, and (4) insufficient ecological and distribution data for many of the species. Three species that have been reported from the Pontocaspian region are not included in this list. The bithyniid gastropod Alocinma caspica (Westerlund, 1902) has been described from the east side of the Caspian Sea (Beriozkina et al. 1995 indicated this record is probably from the vicinity of Krasnovodsk, Turkmenistan). However, Starobogatov et al. (2004) argued the species lives in waterbodies of Bol'shoy Balkhan (Turkmenistan) and probably not in the Caspian Sea itself (Vinarski et al. 2013, Vinarski and. Furthermore, two Pseudamnicola species have been described from Lake Razim in Romania (P. leontina Grossu, 1986 andP. razelmiana Grossu, 1986) that is prime Pontocaspian habitat. Like bithyniids, Pseudamnicola has not been reported as a Pontocaspian group elsewhere, and probably they are freshwater species that live in the surrounding streams or in springs. For now, we have excluded these species from the Pontocaspian species list.
This list contains 55 accepted and a further 44 uncertain endemic Pontocaspian mollusc species (Table 2), here defined as species that are considered to be endemic for at least one of the Pontocaspian basins. There are 14 native and three immigrant species (at least in one of the Pontocaspian basins), even though some species may be native or endemic in one of the basins and have become invasive in another of the Pontocaspian Basins. All species that have an uncertain status belong to the Pontocaspian category. The Caspian Sea Basin has the highest number of accepted endemic Pontocaspian species (48) but also poses the greatest taxonomic challenges, with a further 37 species whose status are unclear.
The species richness estimate reflects the current shift of molluscan systematics from morphology-based to integrated studies, with increasing contributions of molecular and statistical species delineation approaches . It has recently been shown that many nominal taxa of fresh-and brackish-water snails and mussels described on the basis of their shell characters (the Pontocaspian molluscs rarely were described on the base of anatomical studies) lack a genetic support (with few exceptions such as e.g., Popa et al. 2012, Stepien et al. 2013) and thus do not represent evolutionary meaningful units. On the other hand, cryptic speciation is known within many taxa of molluscs in long-lived lakes (Albrecht et al. 2006), and the Pontocaspian biota may include some previously unrecognised species. Thus, we consider our check-list rather as a starting point for further integrated research, not a definitive and fixed inventorisation of the Pontocaspian molluscs.
Anyone who reads this list or works such as Logvinenko and Starobogatov (1969) or Vinarski and Kantor (2016) may think that the Caspian Sea still maintains its unique and species-rich mollusc fauna. However, the actual state of affairs is problematic as many species thought to be endemic to this large saline lake have not been found since their description, and recent attempts to obtain fresh material for genetic studies mostly failed. Clearly, the conservation status of Pontocaspian species is insufficiently known. With our working list we aim to assist in the necessary follow-up conservation assessments.  (Linnaeus, 1758) A N Theodoxus pallasi Lindholm, 1924* A N N N/EX? Theodoxus schultzii * U E Theodoxus velox V. Anistratenko in O. Anistratenko et al., 1999 A N Eupaludestrina stagnorum (Gmelin, 1791) A N/IM N/IM Caspia baerii Clessin & Dybowski in Dybowski, 1887 A E? E ?Caspia valkanovi (Golikov & Starobogatov, 1966) U E Clathrocaspia brotzkajae (Starobogatov in Anistratenko & Prisjazhnjuk, 1992) A ?E E Clathrocaspia gmelinii (Clessin & Dybowski in Dybowski, 1887) A E Clathrocaspia isseli  U E

Species
Status BSB CSB AS Turricaspia uralensis  U E Turricaspia vinogradovi  U E Abeskunus brusinianus (Clessin & Dybowski in Dybowski, 1887) A E Abeskunus depressispira  A E Abeskunus exiguus  A E Andrusovia andrusovi Starobogatov, 2000 U E Andrusovia brusinai Starobogatov, 2000 U E Andrusovia dybowskii Brusina in Westerlund, 1902 A E Andrusovia marina  U E Lithoglyphus naticoides (Pfeiffer, 1828)   U E Gyraulus sulcatus  U E Most taxonomic difficulties were encountered for the bivalve genera Monodacna and Dreissena and the Pyrgulinae gastropods (especially genera Turricaspia and Laevicaspia). Furthermore, there is an urgent need to assess whether representatives of species complexes in the three main Pontocaspian basins (Aral Sea, Caspian Sea, Black Sea) concern separate species as several of these regional populations are in immediate danger of extinction or already extinct (for example with the disappearance of the Aral Sea). Combined methodological efforts will enable us to estimate the extent and characterise the nature of Pontocaspian faunal turnover, and this species list is a first attempt in the required uniform taxonomic base.