A late Pleistocene gastropod fauna from the northern Caspian Sea with implications for Pontocaspian gastropod taxonomy

Abstract The present paper details a very diverse non-marine gastropod fauna retrieved from Caspian Pleistocene deposits along the Volga River north of Astrakhan (Russia). During time of deposition (early Late Pleistocene, late Khazarian regional substage), the area was situated in shallow water of the greatly expanded Caspian Sea. The fauna contains 24 species, of which 16 are endemic to the Pontocaspian region and 15 to the Caspian Sea. The majority of the species (13) belongs to the Pyrgulinae (Hydrobiidae), a group famous for its huge morphological variability in the Pontocaspian region. The phenotypic diversity has led to an inflation of genus and species names in the literature. New concepts are proposed for many of the genera and species found in the present material, with implications for the systematics and taxonomy of the entire Pontocaspian gastropod fauna. Laevicaspia vinarskii sp. n. is described as a new species. This contribution is considered a first step in revising the Pontocaspian gastropod fauna.


Introduction
The Caspian Sea is Earth's largest inland water body. With an area of 378,100 km² it covers about 40% of the world's continental surface water (Dumont 1998, Lehner andDöll 2004). The endorheic Caspian Basin is situated at the crossroads between Europe and Asia and borders Azerbaijan, Iran, Kazakhstan, Russia, and Turkmenistan. Today, its water balance is strongly controlled by the rivers Volga (Russia) and Ural (Kazakhstan) entering from the north and the Kura River (Azerbaijan) flowing in from the southwest and by evaporation from the sea and the adjacent Kara Bogaz Gol (Dumont 1998). The Caspian Sea is a mesohaline lake with an average salinity of about 12.8‰. Steep salinity gradients exist in the northern Caspian Sea from near freshwater conditions at the Volga River delta in the north to a maximum of 13.8‰ in the southeast (Dumont 1998).
During the Pleistocene, several major transgressive-regressive cycles caused recurrent connections between Black Sea and Caspian basins, which were accompanied by dramatic changes in lake size, salinity and biotic assemblages (e.g., Dumont 1998, Yanko-Hombach et al. 2007, Svitoch 2008, Shkatova 2010, Forte and Cowgill 2013, Van Baak et al. 2013, Yanina 2013, 2014, Taviani et al. 2014. In spite of the major environmental fluctuations over the geological past, the Caspian Sea Basin hosted a succession of anomalohaline to freshwater lakes since the late Pontian (late Messinian, late Miocene; Popov et al. 2006, Van Baak et al. 2013. The extensive duration facilitated the accumulation of diverse and highly endemic ("Pontocaspian") biota in this long-lived lake (sensu Gorthner 1994), especially since the Early Pleistocene. As to the recent mollusk fauna, 92 species of gastropods and 35 species of bivalves are listed in latest systematic catalogues (e.g., Kantor and Sysoev 2005, 2006, Kantor et al. 2010, Vinarski and Kantor 2016. As for the gastropods, which are dominated by small-sized Hydrobiidae, 92.4% of them are endemic to the Caspian Sea (Neubauer et al. 2016a). Because of its high diversity, the Caspian Sea has been classified as a major biodiversity hotspot for anomalohaline gastropods (Neubauer et al. 2015a). However, the endemic mollusk fauna is at present severely suffering from the expansion of a number of invasive species (Kosarev and Yablonskaya 1994, Grigorovich et al. 2003, Orlova et al. 2005, Therriault et al. 2004, Riedel et al. 2006, Heiler et al. 2010, Albrecht et al. 2014. Since the early 20 th century, human activity has led to a massive increase in the rate of establishment of non-indigenous aquatic species compared to preceding natural colonization (Grigorovich et al. 2003). Additional environmental pressure is exerted on the resident fauna by the increasing concentrations of heavy metals and pesticides (e.g., Agusa et al. 2004, Anan et al. 2005. In order to predict future biodiversity loss as a response to natural or anthropogenically induced environmental change, it is vital to document and understand the species richness and development of the endemic fauna over longer temporal scales. For this purpose, a sound taxonomic framework is required. The extreme morphological variability of many of the described species complicates taxonomy and, thereby, hampers reliably diversity assessments. Preceding taxonomic studies carried out in the 19 th and 20 th century have produced a plethora of available species names, partly based on minor morphological deviations. Taxonomic works are hampered by (1) the inadequate nature of descriptions and illustrations, (2) the apparent loss of much of the material, (3) the few and hugely variable morphological characters in some of the groups, and (4) the apparent recent loss of many of the species, which makes combined morphological and molecular approaches impossible. Presently, the statuses of most Caspian endemic gastropods, especially of the numerous representatives of the Pyrgulinae (Hydrobiidae), are poorly resolved.
The present contribution details a diverse gastropod fauna from upper Khazarian (Upper Pleistocene) deposits from the northwestern part of the Caspian Basin, at that time witnessing a major transgressive event (Svitoch 2012; Fig. 1). We provide descriptions, illustrations and comparisons of the so far mostly poorly known species, and suggest nomenclatural and taxonomic rectifications. Since we could examine little of the type material of the discussed species (mostly because the whereabouts are  Yanina (2014), who suggested an absolute lake level of 10 m b.s.l. at that time. Considering the present Caspian base level of 27 m b.s.l., this estimate corresponds to a lake level rise of 17 m. (Note that the model is restricted to the Pontocaspian catchment area and disregards potential topographic differences.) The bathymetry ranges are based on the GEBCO_2014 model (version 20150318) for present-day (Weatherall et al. 2015); shown isobaths equal to 100, 500 and 1000 m below current lake level. unknown), we limit our conclusions on former concepts and potential synonymies to taxa that have been thoroughly described and/or adequately illustrated (e.g., Kantor and Sysoev 2006). One particular focus of the present work is the revision of genus concepts that have been applied to Pontocaspian Hydrobiidae.

Materials and methods
The studied mollusk fauna derives from deposits exposed near the small village of Selitrennoye (also as Selitrennoe; Russ. Селитренное) along the left bank of the Akhtuba River, a distributary of the Volga River (Russia) (Fig. 1). The locality is situated about 100 km NNW of the city of Astrakhan in the administrative division of the same name (47°10'21.19"N,47°26'25.41"E,WGS84). The investigated section of 14 m height spans the upper Khazarian to Khvalynian regional substages, which correlates to the early Late Pleistocene (Svitoch 2012, Yanina 2013, 2014. The base of the Quaternary outcrop, which lies 19 m below sea level, is formed by 2.5 m of upper Khazarian sands with common dispersed shells, including shell lenses (Fig. 2). This layer contains the here described gastropod fauna and several species of Lymnocardiinae bivalves. Upsection follows a 1-m-thick interval of horizontally alternating sandy and silty layers containing a diverse assemblage of bivalves of the genera Monodacna, Didacna, and Dreissena. Above it, 4 m of clays containing siltstones and sand layers were deposited. Overlying the interval, 3 m of lower Khvalynian sands are present, containing species of Didacna and Dreissena, followed by 1 m of brown silty clays ("chocolate clays"). The top of the Pleistocene deposits is marked by 1 m of upper Khvalynian sands and sandy loams barren of fossils, topped by a late Holocene soil complex rich in archeological remains.
Approximately 5 kg of sediment were collected by F.W. in September 2015 and were washed over a 0.5 mm sieve before sorting. All material is stored at the Faculty of Geography of the Moscow State University under collection numbers LV 201501-201530 and 201731-201750 and at the Naturalis Biodiversity Center, Leiden, The Netherlands, under collection numbers RGM 1309784-1309793, 1309797-1309856, 1310190-1310249, and 1310252-1310258. Macro-photographs of the specimens were taken with a Leica M165 C stereomicroscope with attached DFC420 camera, using the focus stacking function of the Leica Application Suite software v. 4.4.0 at the Naturalis Biodiversity Center, Leiden. SEM images were acquired on a JEOL JSM-6480LV at the same institute. Specimens were coated with a 20 nm thick platinum-palladium alloy in a Quorum Q150T S coater.
For every species, a number of specimens was measured as representatives of its morphological spectrum. Shell measurements for Theodoxus are given as height × largest width (perpendicular to height) × second-largest width (perpendicular to both other axes); for all other species, measurements are given as height × width. Counting of protoconch whorls follows the method used by Verduin (1977) (Fig. 3). Descriptions and information on the whereabouts of type material are only indicated for Pontocaspian species; a brief account on the non-indigenous species detected herein is provided   Verduin (1977). at the end of the Systematic Paleontology section. Synonymy lists comprise original descriptions, records providing illustrations and entries in systematic catalogues referring to Caspian records (e.g., Kantor and Sysoev 2006, Vinarski andKantor 2016). The systematic classification follows Bouchet et al. (2017) and MolluscaBase (2017).
Abbreviations used are:

Systematic paleontology
We tried to locate the depository of type specimens for all identified species, but this was successful only for a part of the fauna. In particular, much of the type material of the species described by Logvinenko and Starobogatov (1969) could not be located, since these authors did not provide information on the depository of the types or the localities they were retrieved at. According to Kantor et al. (2010), the types should be stored in ZIN but few have been found, since large parts of Starobogatov's collection have not been entirely inventoried as yet. Description. Near globular shell with up to 2.7 whorls. Protoconch consists of about half a whorl; diameter of about 570 µm; nucleus measures ca. 250 µm in diameter; surface mostly corroded; P/T transition indistinct, marked by onset of growth lines. Apex weakly raised. Last whorl passes from upper suture over weakly inclined ramp with shallow concavity into broadly, regularly rounded flank that is near semicircular in profile; relative length of ramp increases with ontogeny. Aperture inclined, regularly semicircular. Callus moderately thickened, glossy, edentate; right margin bulging, symmetrically sinuate, with near straight-sided lower and upper thirds and broad, shallow indentation in central third; left margin extends sinuate over base of penultimate whorl, with small adapical indentation, formed by slightly protruding peristome margin. Peristome sharply edged throughout ontogeny from adapical tip to where it passes into callus margin at base of penultimate whorl. Adapically, peristome margin forms steep crest towards callus, sometimes accompanied by thin, shallow furrow at the transition. Color pattern already starts on early teleoconch as widely spaced, dark yellow to brown curved lines, which pass into slightly irregular zigzag lines with partly dichotomizing branches on last whorl; line width, density, amplitude, color and raggedness varies among specimens and partly within the same individual.

Class
Discussion. The regular, widely spaced zigzag pattern is characteristic of the species. Comparable patterns occur in T. danubialis (Pfeiffer, 1828) and occasionally in T. fluviatilis (Linnaeus, 1758), but in these species lines are finer and more closely spaced. They furthermore differ in their less elongated shells. Similarly, T. euxinus (Clessin, 1886) from the Black Sea is more globular and shows a much denser and finer color pattern (Kantor and Sysoev 2006, Welter-Schultes 2012. Theodoxus schultzii (Grimm, 1877) has traditionally been distinguished from T. pallasi by its rounder shell and the massively expanded aperture (Zettler 2007). Currently, the whole group is under study using genetic data. Preliminary results suggest that both T. pallasi and T. schultzii may be grouped with the Armenian species T. major Issel, 1865, and possibly a major name change for T. pallasi is due (A.F. Sands, pers. commun. 05/2018).
Theodoxus astrachanicus Starobogatov in Starobogatov et al., 1994 from the Azov Sea and Volga delta is claimed to differ from T. pallasi in size and rate of whorl expansion . However, both species correspond well in terms of shell shape and, in particular, the typical zigzag pattern (see also Kantor and Sysoev 2006). We therefore agree with Anistratenko et al. (2017) to treat T. astrachanicus as a junior synonym of T. pallasi.
Distribution. Presently living in the Caspian Sea, the Sea of Azov and the Aral Sea; records from Armenia and the Ural River need confirmation (Anistratenko et al. 2017). In the Pleistocene, the species also dwelled in river deltas entering the Black Sea, where it probably became extinct during the Neoeuxinian/late Pleistocene (Anistratenko et al. 2017). Discussion. The genus Caspia has been widely used for species with small ovoid shells, occasionally with spiral or reticulate teleoconch sculpture. Based on the expression of sculpture, some authors have divided the species among the (sub)genera Caspia s.s., with a single spiral line below the suture, and Clathrocaspia Lindholm, 1930, exposing a reticulate pattern (e.g., Anistratenko and Prisyazhniuk 1992, Boeters et al. 2015, Büyükmeriç and Wesselingh 2018. Species lacking teleoconch sculpture were grouped under the new taxon Ulskia by Logvinenko and Starobogatov (1969). While those authors considered it a subgenus of Pyrgula, W. Dybowski (1887) originally treated its type species (Caspia ulskii Clessin & W. Dybowski in W. Dybowski, 1887, see below) as a sculpture-less form of Caspia.

Subclass
Ulskia ulskii is available in the present material, and we have investigated the type species of Clathrocaspia (Caspia pallasii Clessin & W. Dybowski in W. Dybowski, 1887) obtained from Holocene deposits of the northern and southern Caspian Sea. However, the type species of Caspia, Caspia baerii Clessin & W. Dybowski in W. Dybowski, 1887, is unknown to us. The original description suggests that it is similar to Ulskia and Clathrocaspia in terms of size and shape, yet to differ in the presence of a single line below to suture, demarcating a narrow subsutural ramp. All three genera are probably closely related, which is also suggested by the similar protoconchs of Ulskia and Clathrocaspia (pers. obs. T.A.N.). Since Ulskia and Clathrocaspia can be easily distinguished based on the presence of sculpture, we propose to treat them as distinct genera. The status of Caspia remains doubtful until the type species is properly re-investigated.
Description. Slender ovoid shell with up to 4.7 whorls. Protoconch broad, low dome-shaped, comprising 1.25 whorls that measure 365 µm; nucleus is ca. 105 µm wide; protoconch surface finely but strongly malleate; pattern irregular on initial part and only partly present on nucleus; P/T transition marked by thin axial line and slight step in the upper suture. Teleoconch whorls slightly ton-shaped, weakly convex in abapical half and straight-sided or almost so in adapical half, followed by pronounced convexity at upper suture, producing slightly stepped spire. Last whorl attains ca. 61-66%, descends into steep, straight base. Aperture slender ovoid, slightly inclined, with faint adapical notch at contact to penultimate whorl. Peristome slightly thickened and expanded. In lateral view, outer lip exposes marked adapical indentation and very weak abapical indentation; columellar lip straight. Umbilicus narrow but always open. Growth lines weak but distinctly sigmoidal, with opisthocyrt upper half and prosocline lower half. In addition, faint spiral threads are visible on some shells.
Discussion. Pyrgula schorygini Logvinenko &Starobogatov, 1969 andP. nana Logvinenko &Starobogatov, 1969, both of which were also originally included in the subgenus Ulskia, closely resemble this species. Logvinenko and Starobogatov (1969) did not discuss similarities or differences among the species involved, but their descriptions suggest they considered minor differences in whorl profile and suture depth sufficient to discriminate species. A similar range of variability is present in our sample as well and might rather reflect intraspecific variation. We thus consider the three species synonymous.
Distribution. Endemic to the Caspian Sea, reported from water depths between 45 and 120 m (Logvinenko and Starobogatov 1969).
Discussion. The subfamily placement of the genus follows Vinarski and Kantor (2016: 214) and is based on the resemblance with species of the genus Horatia Bourguignat, 1887 (see also discussion in Starobogatov 2000). A recent molecular phylogeny including the Hydrobiidae suggests the Horatiinae to be distinct from the Belgrandiinae (Wilke et al. 2013; see also Bank 2017). We follow Starobogatov (2000) and regard Caspiohoratia Logvinenko & Starobogatov, 1969 as a junior synonym of Andrusovia. Description. Shell broad trochiform, about as high as wide, with up to 4 whorls. Rarely specimens with slightly elevated spire occur. Protoconch high domical, about semicircular in profile; initial part immersed; consists of 1.1 whorls, measures 300 µm in diameter; nucleus about 90 µm wide; protoconch surface finely but strongly malleate near lower suture, rest appears to be irregularly granulate, but that might be due to poor preservation; P/T boundary sharp, marked by massive growth constrictions near lower suture. Teleoconch whorls highly convex, with maximum convexity in adapical half, producing slightly stepped spire. Last whorl attains 74-81% of shell height. Aperture broadly drop-shaped, slightly inclined, with faint adapical notch at contact to penultimate whorl. Peristome slightly thickened and expanded at columella and base; sinuate in lateral view, with weakly protruding central part and weak adapical indentation. Umbilicus wide, deep. Fine prosocline growth lines cover shell. On one specimen, traces of spiral threads occur on base.

Andrusovia brusinai Starobogatov, 2000
Discussion. The Caspian congeners Andrusovia dybowskii Brusina in Westerlund, 1902a(sensu Starobogatov 2000 and A. andrusovi Starobogatov, 2000 differ from the present species in their much lower spires. Andrusovia marina (Logvinenko & Starobogatov, 1969) is smaller and has a shorter spire. Starobogatov (2000) based the distinction from A. brusinai on minor differences in shell ratios but these are strongly affected by the varying number of whorls and shell size; it might well be that A. marina and A. brusinai are just different growth stages of the same species. Since we have not seen the type material of Logvinenko and Starobogatov (1969), we tentatively accept the distinction of both taxa by Starobogatov (2000). Further comparison with the Logvinenko and Starobogatov material is essential to assess whether the two names refer indeed to distinct species.
Distribution. Endemic to the Caspian Sea, reported from middle and south Caspian Sea at depths between 47 and 311 m (Starobogatov 2000). Description. Shell shape highly variable, ranging from broad ovoid to slender conical, comprising up to 6.5 whorls. Protoconch consisting of about one whorl, with nucleus immersed; initial part slightly raised, producing acute apex; surface weakly granular to malleate; P/T transition clear. Protoconch and teleoconch whorls highly convex, sometimes slightly flattened centrally in later whorls; suture deep. Size of last whorl varies between 55-62%, descends into straight-sided base. Aperture regularly ovoid, slightly inclined, touching base of penultimate whorl, leaving wide umbilicus. Peristome simple, sometimes weakly expanded. Surface smooth expect for very fine prosocline growth lines.

Subfamily
Discussion. The shells of Ecrobia can only be reliably identified on the specieslevel using molecular data (Haase et al. 2010). Therefore, we tentatively assign the detected specimens to Ecrobia grimmi, which is the only Ecrobia species occurring in the Caspian Sea today (Haase et al. 2010).
Most of the species presently assigned to Caspiohydrobia Starobogatov, 1970, including its type species, Pyrgohydrobia eichwaldiana Golikov & Starobogatov, 1966, range within the morphological variability of this species. Previous examination of both reproductive systems (Sitnikova et al. 1992) and juvenile shells (Filippov and Riedel 2009) did not yield criteria supporting interspecific differentiation. Very likely all of the thirty Caspiohydrobia species listed by Kantor and Sysoev (2006) are morphotypes of a single species, probably E. grimmi. Given the problems of using shell morphology to identify Ecrobia, taxonomic conclusions on the synonymy of the Caspiohydrobia species require molecular data.
Note on species authority. W. Dybowski (1887: 7) noted that all diagnoses were drafted by Clessin and himself and most new species were therefore marked with "nob." (Lat. nobis, "us"). However, W. Dybowski obviously made exceptions. In case of the new genus Clessinia, he marked the authority with "m." (Lat. meus, "mine"). For Hydrobia grimmi, the authority is clearly indicated with "Cless.", making Clessin the sole author of the species (unlike indicated by several authors).
Discussion. The Caspian Pyrgulinae (sensu lato) encompasses 64 species that are considered accepted in the current literature (Vinarski and Kantor 2016). However, most of them are poorly known, documented by insufficient descriptions and drawings; for many, the type material has not been found (Kantor and Sysoev 2006, Vinarski andKantor 2016). The extreme morphological variability of several representatives, such as those detected in the material from Selitrennoye, led previous authors to introduce numerous species based on shells with only minor deviations in shape, size or whorl outline. The Caspian Pyrgulinae therefore requires careful revision using molecular and anatomical data as far as available.
In addition to the problems associated with distinguishing species, genus-level classification is poorly resolved as well. Several attempts have been made to categorize this vast variability, and genus concepts have changed tremendously (e.g., B. Dybowski and , Zhadin 1952, Logvinenko and Starobogatov 1969, Kantor and Sysoev 2006, Vinarski and Kantor 2016. Twelve genus names have been described for members of the Caspian Pyrgulinae, based on quite different concepts of traits considered diagnostic. Currently, all species are classified within Caspia Clessin & W. Dybowski, 1887, Pyrgula De Cristofori & Jan, 1832and Turricaspia B. Dybowski & Grochmalicki, 1915(Kantor and Sysoev 2006, Vinarski and Kantor 2016. This scheme unites quite a variety of different morphologies under the same genus names, while at the same time similar species are assigned to different genera (e.g., Kantor and Sysoev 2006). Unfortunately, hardly any previous study provided explanations for their genus classifications or systematic concepts in general.
A thorough revision of all Caspian Pyrgulinae is beyond the scope of this study, but we discuss and revise the concepts that have been applied to the species studied herein. Vinarski and Kantor (2016) listed 38 species of the genus Pyrgula for the Caspian Sea. The type species of Pyrgula De Cristofori & Jan, 1832, P. annulata (Linnaeus, 1758), lives in freshwater lakes and springs in Italy and Dalmatia (Welter- Schultes 2012). Shell morphology, anatomy and protoconch characteristics are very similar to Pontocaspian Pyrgulinae, e.g., some species of Turricaspia (compare also discussion in Riedel et al. 2001). However, molecular evidence suggests that Pyrgula annulata is only distantly related to the Pontocaspian species flock within the Pyrgulinae, with the last common ancestor dating back to the late Miocene (Wilke et al. 2007). Therefore, Pontocaspian species should not be attributed to Pyrgula, despite apparent morphological congruence, especially of some of the keeled Pontocaspian Pyrgulinae. A separation on subfamily level as proposed by B. Dybowski and Grochmalicki (1915) is opposed by the latest phylogeny of rissooidean gastropods, which suggests a rather close relationship (Wilke et al. 2013).
Based on a review of the Pontocaspian species formerly attributed to these genera and illustrated in the literature (Golikov and Starobogatov 1966, Logvinenko and Starobogatov 1969, Alexenko and Starobogatov 1987, Kantor and Sysoev 2006, Anistratenko 2008, we propose to distinguish the genera Clessiniola and Laevicaspia from Turricaspia, and to treat Caspiella and Euxinipyrgula as junior synonyms of Laevicaspia. Clessiniola species can be easily distinguished from species attributed to other genera based on their broad shells with a large body whorl and aperture. The situation for the Laevicaspia-Caspiella-Euxinipyrgula is more difficult. The three type species (see above) share the ovoid shape with cyrtoconoid spire, the high whorl accretion rate, the shape, inclination, lateral sinuation and thickening of the aperture, and the extent and sculpture of the protoconch (e.g., Kantor and Sysoev 2006, Anistratenko 2008, and this study). The only differences are shell size and whorl convexity, which we do not consider sufficient to distinguish genera. The adapical thickening of the aperture resulting from downward growing of the shell in late ontogeny as stated in the diagnosis of the genus Euxinipyrgula by Sitnikova and Starobogatov (1999) is also shown for species that have been attributed to Caspiella (see below). The features of the soft-part anatomy considered diagnostic by these authors need to be rechecked and compared to live material from the Caspian Sea to reevaluate the position of Euxinipyrgula. Sitnikova and Starobogatov (1999) also discussed the similarities between Caspiella and Euxinipyrgula, concluding that Caspiella should perhaps be included in the genus Euxinipyrgula, possibly as a separate subgenus (which would be nomenclaturally invalid however).
The ovoid shape, lateral sinuation and thickening of the aperture typical for the Laevicaspia-Caspiella-Euxinipyrgula group are also found among species of the genus Prososthenia Neumayr, 1969 from the middle Miocene of the Dinaride Lake System (e.g., Neubauer et al. 2016b). These species, however, differ in the granulate protoconch making up less than one whorl.
Species of Turricaspia differ from Laevicaspia in the slower, regular whorl accretion, producing a conical spire and a higher number of whorls at the same size. In addition, Turricaspia species have usually more fragile shells, thinner peristomes and often more strongly sinuate growth lines.
The genus Caspia is listed among Pyrgulinae in latest catalogues (Kantor and Sysoev 2006, Vinarski andKantor 2016), but it has been shown to be unrelated to that subfamily , Bouchet et al. 2017; see discussion of the Caspiinae above).
Finally, several Pontocaspian Pyrgulinae have been previously assigned to the genus Micromelania Brusina, 1874 (e.g., W. Dybowski 1887, B. Dybowski andGrochmalicki 1917). Its type species, Micromelania cerithiopsis Brusina, 1874 (subsequent designation by Dollfus 1912), derives from late Miocene deposits of Lake Pannon. It differs considerably from Pontocaspian Pyrgulinae regarding the presence of 2-4 noded keels and the small size (4.5 × 1.33 mm after Brusina 1874) compared to the rather high number of eight whorls.  Description. Broadly drop-shaped to rarely conical shell of up to six whorls. Protoconch insufficiently preserved to specify extent and surface sculpture; P/T transition indistinct; first whorl measures ca. 340 µm in diameter. Teleoconch whorls moderately and regularly convex; sometimes, spire is very faintly stepped; suture narrow. In many specimens, shells starts to grow stronger in abapical direction in course of last (two) whorl(s), producing non-parallel suture and relatively higher penultimate whorl. Rarely, forms with comparatively slender shape and regularly increasing whorls (and thus relatively smaller last and penultimate whorls) occur. Both types are linked via intermediates. Aperture regularly ovoid, inclined; inner lip glossy, weakly to sometimes more prominently thickened; strongly adnate, sheet-like expanded over base of penultimate whorl and columella, rarely leaving very narrow umbilicus; broad, shallow spout occurs at transition between columella and base; outer lip mainly thin, sometimes weakly thickened at anterior notch. Growth lines very faint, with prosocline upper third and near orthocline lower two-thirds.

Genus
Discussion. This species displays a large morphological variability within our ample material. Shell shape ranges between slender conical to broadly ovoid, sometimes with weakly irregular growth. Likewise, shell size, whorl convexity, and number of whorls vary considerably. Yet, these features intergrade without clear boundary, rendering a distinction of species unreasonable.
The morphological variability is not restricted to our material but a general feature of Clessiniola. It was documented by several previous authors, partly for specimens from the same localities (e.g., Eichwald 1838, Issel 1865, W. Dybowski 1887-1888, Golikov and Starobogatov 1966, Logvinenko and Starobogatov 1969, Alexenko and Starobogatov 1987, Anistratenko et al. 2011. The species concepts applied by the different authors, however, varied greatly. The present material includes shells that have been variably attributed to the species C. variabilis (Eichwald, 1838), C. triton (Eichwald, 1838) and C. martensii (Clessin & W. Dybowski in W. Dybowski, 1887). The triton-morphotype sensu Eichwald characterizes broad specimens with slightly detached aperture (see also Kantor and Sysoev 2006); these forms are rarely represented in our material. Eichwald (1838Eichwald ( , 1841 himself confirmed the rarity of the form, also stating that he did not find a living representative (in contrast to C. variabilis). The martensii-morphotype was introduced for similarly broad morphologies. (Note that Clessin and W. Dybowski used a different concept of C. triton, there having a rather elongate conical shell.) Because of the fluent morphological transition between forms traditionally referred to as C. variabilis, C. triton and C. martensii, as well as their joint occurrence in several localities in the Pontocaspian region, one might consider all of them synonymous. Personal observations on Holocene material from Dagestan area, however, indicate indeed distinguishable morphotypes without intermediates. Moreover, frequent shell repair found in most of the Selitrennoye specimens additionally complicates an unbiased view on morphological diversity. A more in-depth investigation comparing undamaged material from different sites is thus required.
Clessinia ahngeri Westerlund, 1902 is often listed as junior synonym of C. variabilis, but without discussion (e.g., Logvinenko andStarobogatov 1969, Vinarski andKantor 2016). The original description of C. ahngeri suggests close similarities indeed between both species claiming, however, that it differs from other congeners in the much larger spire (11 × 5 mm) and the slightly sinuate outer lip. Examination of Westerlund's (1902b) material is required to ascertain the alleged synonymy.
The record of "Paludina Eichwaldi Kryn." Eichwald (1841) listed in synonymy of C. variabilis refers to a nomen nudum mentioned in a species list by Krynicki (1837).
Distribution. Endemic to the Pontocaspian region. Found in the Caspian Sea and the lower courses of rivers and freshwater parts of the Azov and Black seas (Anistratenko et al. 2011, Vinarski andKantor 2016). Also reported from Neoeuxinian (late Pleistocene) deposits of the Marmara Sea (Taviani et al. 2014).  Logvinenko and Starobogatov (1969). Caspian Sea, Pleistocene.

Genus
Discussion. Lindholm (1922) studied the type material of Buliminus goebeli Westerlund, 1896 from Mangyschlak (Mangystau Peninsula, Kazakhstan) and concluded that is a junior synonym of "Micromelania" caspia (Eichwald, 1838). Westerlund (1902a), considering Buliminus goebeli as a member of terrestrial "Bulimoidea" (= Enidae), introduced the new genus Thaumasia, which would take precedence over Laevicaspia B. Dybowski & Grochmalicki, 1917. However, Thaumasia Westerlund, 1902 is invalid as a junior homonym of Thaumasia Perty, 1833 (Arachnida) and Thaumasia Albers, 1850 (Gastropoda, Subulinidae) (see also Lindholm 1925 Alexenko and Starobogatov (1987). Description. Large, slender ovoid shell comprising up to 8.3 whorls. Protoconch large, measuring 535-600 µm at 1.15-1.2 whorls, with initial part inflated; nucleus almost immersed, 190-230 µm wide; nucleus and early protoconch bear intentions of malleate sculpture, which passes into granular surface after half a whorl accompanied by onset of spiral striae; P/T boundary indistinct. Whorl convexity decreasing rapidly: first teleoconch whorl moderately convex, second to last whorl low convex, sometimes almost straight-sided; maximum convexity is in lower half; whorls closely attached, suture narrow; a very small but marked convexity appears at upper suture, producing a faintly stepped spire; occasionally, it is accompanied by shallow abapical concavity. Last whorl makes up 46-50% of shell height, passing over regular but weakly convex to near straight-sided to slightly concave base. Aperture slender ovoid, inclined, closely attached to preceding whorl; in latest ontogeny, shell growth is more abapically directed, resulting in marked thickening at adapical tip. Peristome simple, thin, slightly expanded and indented at base; distinctly and regularly sigmoidal in lateral view, with upper half broadly indented and lower half broadly protruding; inner lip protrudes in lateral view, extending sheet-like over base of penultimate whorl; umbilicus mostly closed, rarely very narrow, slit-like. Growth lines weakly sigmoidal: strongly prosocline in upper half, weakly opisthocline in lower half. Several specimens show faint spiral threads on last and penultimate whorls.
Discussion. Different concepts of this species previously applied have led to considerable confusion about its real identity. This is partly rooted in the description and illustration provided by Eichwald (1838Eichwald ( , 1841 that were insufficient to allow safe discrimination from similar species. For instance, Micromelania caspia sensu Grimm (1876Grimm ( , 1877, W. Dybowski (1887)(1888) and B. Dybowski and Grochmalicki (1917) differs from the present species in the acute apex, the moderately convex whorls, the deep suture and the thin peristome. As already noted by Alexenko and Starobogatov (1987), it represents a different species, i.e., Laevicaspia lincta (Milashevich, 1908). That species was described from Lake Katlabukh near the Danube delta in Ukraine (lectotype, which matches Milashevich's description, is illustrated in Kantor and Sysoev 2006: 95, pl. 45, fig. D;as Euxinipyrgula lincta). Specimens from the Neoeuxinian (late Pleistocene) of the Marmara Sea identified as E. cf. lincta by Taviani et al. (2014) differ from that species in the near straight-sided whorls and thickened peristome; in fact, the material corresponds well to L. caspia.
Micromelania caspia sensu Nalivkin (1914) comprises at least two species, both being more elongate, having more whorls and relatively smaller last whorls than L. caspia. In turn, some of the illustrated syntypes of "Micromelania" curta Nalivkin, 1914 and the variety "Micromelania" curta var. planoconvexa Nalivkin, 1914 from Bakunian deposits of Shikhovo, Apsheron Peninsula, Azerbaijan, closely resemble the present spe-cies and are thus (partly) considered synonymous. "Micromelania" curta encompasses a great variability of shapes, ranging from slender, elongate (caspia-type) to broad, conical shells. Since no holotype or lectotype have been designated, the status of this species is unresolved at present. Note that Pyrgula curta sensu Logvinenko and Starobogatov (1969) and Kantor and Sysoev (2006) does not correspond to Nalivkin's species but rather to the specimens Nalivkin (1914) misidentified as Micromelania caspia.
Similarly, Pyrgula caspia sensu Logvinenko and Starobogatov (1969) is a quite different species, showing highly convex whorls and an inflated last whorl. It rather ranges within the morphological variability of Turricaspia meneghiniana (see below). Alexenko and Starobogatov (1987) finally brought stability to the identity of L. caspia by designating a lectotype (see Kantor and Sysoev 2006: 106, pl. 49, fig. A; as Turricaspia caspia), which matches well our specimens. The label accompanying their specimen reads "Kaspiyskoye more" ("Caspian Sea"), which differs from the information provided by Eichwald (Dagestan) (see also discussion in Vinarski and Kantor 2016: 246). Inspection of the catalogue of the Zoological Institute, Russian Academy of Sciences, St. Petersburg (ZIN), however, confirmed that the lectotype is based on Eichwald's original material.
The similar Laevicaspia iljinae (Golikov & Starobogatov, 1966) from Holocene deposits of the Crimean Peninsula can be distinguished in its more slender shape and the spruce-like whorl outline (i.e., steep, straight-sided upper two-thirds passing over convexity into flatter, convex lower third; see also Kantor and Sysoev 2006: 108, pl. 49, fig. D).
Distribution. Endemic to the Caspian Sea (Logvinenko and Starobogatov 1969 stated that the species occurs at a depth of 30-150 m in the middle and southern Caspian Sea, but these data have to be revised given their incorrect concept of L. caspia).
Laevicaspia cincta (Abich, 1859) comb. n. Description. Slender ovoid shell with up to 6.5 whorls. Protoconch broad, low dome-shaped, consists of 1.2 whorls that measure 415 µm in diameter, with slightly inflated initial part; nucleus 150 µm wide; protoconch surface weakly granulate, with intentions of striae on second half; P/T transition distinct, formed by sharp, thin axial line. Whorl convexity decreases steadily during ontogeny, with early teleoconch whorls being moderately convex and penultimate and last whorl low convex to almost straight-sided. On third teleoconch whorl, weak subsutural band emerges that slightly enhances during ontogeny; band forms weak bulge throughout, with maximum convexity in its lower half and steep, almost straight-sided ramp in upper half; abapical demarcation clear, sometimes accompanied by thin groove. Last whorl attains 54-65% of shell height, passing from flattened whorl flank over marked convexity into steep, straight-sided base. Aperture near drop-shaped, inclined, with acute adapical angle, straight parietal margin, obtuse angle between parietal and columellar margins, sometimes slightly expanded palatal margin. Peristome not thickened, weakly expanded at columella and base; regularly sinuate in lateral view, with broad adapical indentation and about equally broad and high abapical protrusion. Umbilicus closed or very narrow. Growth lines weakly prosocline in upper half, near orthocline in lower half.
Discussion. The Selitrennoye specimens match with the original description in terms of size (shell height: 3-4 mm), the ovoid shell shape, the number of whorls, the rounded last whorl and the simple peristome margin; they differ in the expression of the subsutural band, which Abich indicated to be "weakly keeled". We consider these differences to range within the intraspecific variability of this species.
Laevicaspia cincta can be readily distinguished from other Pontocaspian Pyrgulinae by its ovoid, slightly stepped shell with broad, blunt apex, subsutural band and flattened whorls in later ontogeny. Laevicaspia abichi (Logvinenko & Starobogatov, 1969) from the middle Caspian Sea, differs in the much larger size (6.8 × 3 mm), the conical shape, the narrower subsutural band and the larger aperture. The Caspian endemic species Laevicaspia kowalewskii (Clessin & W. Dybowski in W. Dybowski, 1887) resembles L. cincta very closely in terms of the slender ovoid shape with near straight-sided whorls, the closely attached aperture with thin peristome, and the lacking umbilicus; it differs in the lack of a subsutural band and the more elongate shape.
Caspia orthii Clessin & W. Dybowski in W. Dybowski, 1887 was synonymized with the present species by previous authors (e.g., Logvinenko and Starobogatov 1969, Kantor and Sysoev 2006, Vinarski and Kantor 2016. The original description matches well our specimens in terms of size (4.8 × 1.9 mm), number of whorls, expression of the subsutural band and shape of the aperture; the only difference is the "elongatedconical" shape compared to the ovoid shells of L. cincta described by Abich (1859) and represent by our material. Although not having seen W. Dybowski's type material, we tentatively follow the previous assessment and consider Caspia orthii a junior synonym of Laevicaspia cincta.
Distribution. Endemic to the Caspian Sea, in the southern part at a depth of >250 m (Parr et al. 2007). Fig. 9I Description. Ovoid, glossy shell with up to 6.8 whorls. Shell outline variable, depending on growth stage: shells with up to 5 whorls are rather broad, nearly conical; in late ontogeny, shell growth is directed adapically, producing more elongate shapes with narrow, high last whorl; sometimes, these slender elongate morphotypes have slightly irregular shape. Protoconch consists of 1.2 whorls with 355 µm in diameter; nucleus almost immersed, 125 µm wide; surface faintly malleate or granulate, with intentions of spiral sculpture detected in some specimens; P/T boundary very distinct, marked by sharp, thin axial line. Teleoconch whorls weakly to moderately convex, sometimes adapically flattened. Last whorl attains between 55-63% of total height, grades into straightsided or weakly convex base. Aperture drop-shaped, inclined, closely attached to base of preceding whorl, usually covering or rarely leaving slit-like umbilicus. Peristome slightly expanded, thin or thickened all around, especially at adapical tip; regularly sinuate in lateral view, with broad adapical indentation and about equally broad and high abapical protrusion. Growth lines weak, prosocline in upper half, orthocline in lower half. Logvinenko and Starobogatov (1969) listed "Rissoa conus Eichwald, 1841, partim" in synonymy of Pyrgula kolesnikoviana Logvinenko & Starobogatov in Golikov & Starobogatov, 1966 (now classified in Laevicaspia; see below), but without any explanation. The synonymy list was expanded as "Rissoa conus sensu Eichwald, 1841, partim, non Eichwald, 1838" by Kantor and Sysoev (2006 and Vinarski and Kantor (2016), yet again without discussion. The synonymy is not mentioned in the original description of Laevicaspia kolesnikoviana in Golikov and Starobogatov (1966). Very likely, the synonymy roots in the ambiguous description of Eichwald (1838Eichwald ( , 1841, summarizing two different morphologies. Eichwald referred to the typical form as having a conical shell with seven, gently increasing whorls, whereas the last two are much broader; the size was indicated as 2 × 1 lin., which corresponds to 4.2 × 2.1 mm (given Eichwald used the Russian liniya). In addition, he mentioned rarer, slightly longer (3 lin.) specimens, with deeper suture and straight-sided whorls. In 1841, Eichwald illustrated one of these rare specimens. The description in the 1841-work, however, is almost identical to the original description. In this light, it remains unclear why Kantor and Sysoev (2006) and Vinarski and Kantor (2016) referred to as "Rissoa conus sensu Eichwald, 1841, partim, non Eichwald, 1838" in their synonymy lists of L. kolesnikoviana. To complete confusion, the specimen illustrated in Kantor and Sysoev (2006) is not L. conus, differing in the broad, blunt apex and the near straight-sided whorls; it rather resembles L. kowalewskii (Clessin & W. Dybowski in W. Dybowski, 1887).

Discussion.
The holotype of L. kolesnikoviana illustrated by Kantor and Sysoev (2006, pl. 47, fig. N) corresponds to the description and illustration of the rare, slender morphology of Laevicaspia conus sensu Eichwald in terms of the number of whorls and the near straight-sided whorls; it differs only in the considerably smaller size (3.7 mm vs. 6.3 mm). Yet, Golikov and Starobogatov (1966) and Logvinenko and Starobogatov (1969) indicate larger sizes for L. kolesnikoviana (5.5 mm and 6.5 mm, respectively), suggesting a great variability in size. On the other hand, the two morphologies delineated by Eichwald also match our own observations on L. conus. In late ontogeny, growth is directed almost entirely abapically, resulting in more elongate shells with an additional whorl. These larger morphologies correspond completely to the smaller, relatively bulkier shells in all other aspects, which is why we consider them as morphotypes rather than species-group taxa. Without Eichwald's material at hand it is difficult to arrive at a conclusion on this matter.
"Eulima conus Eichwald" as described and illustrated by Grimm (1876Grimm ( , 1877 has little resemblance to actual L. conus. He illustrated a very elongate, conical shell with many more and almost perfectly straight-sided whorls. This fact was already noticed by Clessin and W. Dybowski a few years later, and they introduced Micromelania grimmi Clessin & W. Dybowski in W. Dybowski, 1887 for the misidentified species. The illustrations of specimens from the Kuyalnikian (late Pliocene to early Pleistocene) of the Odessa region identified as Prososthenia conus by Sinzov (1896) show shells with similar shape, proportions and whorl convexity. A more detailed examination of material from the region is required to assess whether it is indeed conspecific with L. conus.
Distribution. Endemic to the Caspian Sea, reported from depths between 0 and 120 m (Logvinenko and Starobogatov 1969). Golikov & Starobogatov, 1966) comb. n. Fig. 10A fig. 366 (1). 2006Pyrgula kolesnikoviana Logvinenko et Starobogatov in Golikov et Starobogatov, 1966 fig. N Description. Small, slender ovoid, shiny shell with up to 6.9 whorls. Protoconch consists of 1.2 whorls, measuring 355 µm in diameter; nucleus rather long, ca. 130 µm wide; surface finely granulate (maybe due to preservation; traces of finely malleate to irregularly striate pattern occurs on margins of nucleus and initial part); faint striae on last third; P/T boundary distinct. Whorl convexity of teleoconch whorls decreasing: first to second whorl moderately to highly convex, last whorl low to moderately convex. Faint subsutural band appears on later teleoconch whorls in some specimens, sometimes accompanied by weak concavity below. Last whorl attains 50-57% of shell height, passing via broad, regular convexity in to weakly convex base. Aperture ovoid, inclined, closely attached to preceding whorl; in latest ontogeny, shell growth is more abapically directed, resulting in marked thickening at adapical angle. Peristome thin  (Logvinenko & Starobogatov in Golikov & Starobogatov, 1966) or thickened all around, with parietal margin sometimes slightly expanded; weakly but regularly sinuate in lateral view, with broad adapical indentation and about equally broad and high abapical protrusion. Umbilicus usually closed or very narrow, slit-like. Growth lines weak, prosocline in upper half, orthocline in lower half. In addition, faint spiral furrows appear in some specimens.

Laevicaspia kolesnikoviana (Logvinenko & Starobogatov in
Discussion. Co-occurring Laevicaspia vinarskii sp. n. differs in the consistently lower whorl expansion rate at the same size and the smaller aperture. Laevicaspia kowalewskii (Clessin & W. Dybowski in W. Dybowski, 1887) can be distinguished by its broader and larger shell.
Description. Slender ovoid shell with up to 6.5 whorls. Protoconch consists of 1.2 whorls measuring 375 µm; nucleus is 140 µm wide; surface strongly granulate on nucleus, less so on remaining protoconch, striae appear on last 0.25 whorls; P/T transition marked by distinct growth rim. Teleoconch whorls moderately convex, separated by narrow suture; whorls increase slowly in height, with the last attaining 53-57% of shell height, passing into weakly convex base. Weak subsutural band is observed in one specimen. Aperture small, inclined, closely attached to base of preceding whorl, leaving no or slit-like umbilicus. Peristome slightly thickened, especially at adapical tip; regularly sinuate in lateral view, with broad adapical indentation and about equally broad and high abapical protrusion. Distinct spiral furrows occur in well preserved specimens. Growth lines weak, prosocline in upper half, orthocline in lower half.
Discussion. The new species differs from co-occurring Laevicaspia kolesnikoviana in the higher whorl expansion rate at about the same size and the larger aperture. Laevicaspia? ismailensis (Golikov & Starobogatov, 1966) from lakes Yalpug and Kugur-lu in the Danube river delta is more slender and larger (5.6 mm) at the same number of whorls and has a less inclined, rounder aperture (see holotype illustrated by Kantor and Sysoev 2006: pl. 50 , fig. A).

Turricaspia andrussowi (B. Dybowski & Grochmalicki, 1915)
Type locality. Caspian Sea (no locality specified). Description. Available fragments indicate very slender, conical shell. Apex broad, blunt, bulbous. Whorl profile flattened, very weakly spruce-like, with straight-sided upper two-thirds passing over convexity into weakly convex lower third; in addition, broad, flat subsutural band appears, sometimes accompanied by very narrow concavity below. Umbilicus seems fully closed. Aperture not preserved in any specimen. Discussion. The identification of the three spire fragments rests upon the strongly adpressed whorls with very narrow suture and the flattened, spruce-like whorl profile, and the large, bulbous protoconch. Turricaspia eulimellula  has a similarly slender spire with adpressed whorls, but it bears a basal keel and the maximum whorl convexity is around mid-height instead of in the lower third. Turricaspia grimmi (Clessin & W. Dybowski in W. Dybowski, 1887) differs in its perfectly straight-sided, rectangular, very weakly stepped whorl profile (see also B. Dybowski and Grochmalicki 1917, pl. 3, figs 34-35;Kantor and Sysoev 2006, pl. 46, fig. L).
A very similar species is Pyrgula dubia Logvinenko & Starobogatov, 1969 from the middle Caspian Sea, matching the present one in the weakly spruce-like whorl profile; in fact, it might just be a juvenile specimen of T. andrussowi. Similarly, Pyrgula turkmenica Logvinenko & Starobogatov, 1969, from the eastern part of southern Caspian Sea, corresponds to T. andrussowi in the weak subsutural band accompanied by an abapical concavity; it might as well be a juvenile representative of T. andrussowi. Logvinenko and Starobogatov (1969) synonymized without discussion Hydrobia spica sensu Grimm, 1876, Turricaspia elegantula sensu B. , T. brusinae (B. Dybowski & Grochmalicki, 1915, as well as several varieties of T. spica and T. turricula described by B. , with T. andrussowi (see also Kantor and Sysoev 2006, Vinarski andKantor 2016). However, none of these taxa actually resembles T. andrussowi. This species can be well delim-ited from these alleged synonyms in its bulbous protoconch and the characteristic, weakly spruce-like whorl profile. (Note that the drawing provided by Logvinenko and Starobogatov 1969 shows a rather broad shell with acute apex; it has little in common with the lectotype designated by them).
Distribution. Endemic to the Caspian Sea (Logvinenko and Starobogatov 1969 indicated occurrences for the middle and southern Caspian Sea at depths of 25-80 m, but based on a much wider concept of the species).
Dimensions. 4.29 × 1.93 mm. Description. Slender elongate shell with ca. 6.5 whorls preserved. Protoconch granulate, originally perhaps densely malleate. First teleoconch whorl straight-sided in profile, passing into weakly convex outline on 2 nd -3 rd whorl. Between 3 rd and 4 th whorl, broad, blunt central swelling emerges, grading into thin angulation on 5 th whorl; no keel is developed. Whorl portion above swelling/angulation straight-sided, below weakly convex; directly above it, weak concavity is formed locally. Aperture ovoid, strongly adnate, leaving no umbilicus, with thin peristome. Growth lines rather distinct, with prosocline upper half and near orthocline lower half.
Discussion. A single subadult shell containing ca. 6.5 whorls (including the protoconch) is available. Size and number of whorls as well as the centrally placed angulation correspond well to Eichwald's (1838Eichwald's ( , 1841 description and illustration of T. dimidiata. However, the central keel is very weakly expressed in our specimen and it starts not before the fourth whorl, which is why we only tentatively attribute it to this species. Kantor and Sysoev (2006) illustrate a much more elongate specimen with cyrtoconoid spire and more abapically placed keel; it might represent a different species. Turricaspia bakuana (Kolesnikov, 1947), likewise described from Caspian Sea, too has a central keel, but differs in the much more slender shell and consistently strong keel from the second teleoconch whorl onwards (cf. Kantor and Sysoev 2006). Turricaspia basalis  has a broader conical habitus and the keel is placed near the lower suture. The subspecies T. b. laticarinata (Logvinenko & Starobogatov, 1969) only differs from T. basalis in the thickness of the keel and is herein considered a junior synonym of the nominal species.
Distribution. Endemic to the Caspian Sea, reported from middle and south Caspian Sea at depths between 35 and 200 m (Logvinenko and Starobogatov 1969).  Fig. 12J).
Description. Slender elongate shell of up to 9 whorls. Protoconch large, measuring about 485 µm in diameter; it forms bulbous cap on top of shell and comprises 1.25 whorls; surface weakly granulate, with striae on last 0.25 whorls; nucleus low, broad, ca. 170 µm in diameter; P/T transition very distinct, marked by sharp growth cessation. Teleoconch whorls low to moderately convex, often flattened or with straight-sided upper half, which creates spruce-like morphology. Sometimes, very weak and thin bulge appears below suture, producing faintly stepped spire. Most shells bear very low and somewhat irregular spirals, but expression varies considerably concerning its onset (mainly starts on lower whorls), strength (faint traces to distinct but blunt keels) and number of elements (one keel near base to several keels spread across whorl profile). Expression of sculpture varies in most specimens throughout ontogeny, which cre-ates uneven, rugged appearance. Aperture comparatively small, in most cases regularly ovoid and weakly inclined, covering up umbilicus entirely or leaving very thin opening; peristome simple. Growth lines strongly sigmoidal, with prosocline upper third and opisthocline lower two-thirds.
Discussion. This species can be distinguished from its congeners in its large, bulbous protoconch and the typical, somewhat irregular sculpture. It is consistently larger, more massive and on average bears much stronger sculpture than co-occurring T. ? spica. The varieties "Micromelania (Turricaspia) spica var. incisata" and "M. (T.) spica var. striata" introduced by B. Dybowski and Grochmalicki (1915) only differ in the depth of the suture and the expression of the teleoconch sculpture, respectively. Given the variability of these features, we consider both of them synonymous with T. lyrata. Already Logvinenko and Starobogatov (1969) considered incisata and lyrata synonymous and, as first revisers, chose lyrata as the valid name of the species. The variety "M. (T.) spica var. lordosa" B.  might also be a synonym of this species. However, the apex of the specimen illustrated in B. Dybowski andGrochmalicki (1915, 1917), which contains diagnostic characters, is not preserved. Nevertheless, striata and lordosa are certainly not synonymous with T. andrussowi as suggested by Logvinenko and Starobogatov (1969). That species differs from T. lyrata in the much slender whorls with sprucelike, near straight-sided profile.
Type material. Not traced. Type locality. "Nei giacimenti fossiliferi di Baku" (from fossil deposits in Baku). Description. Conical shell with up to 9.3 whorls. Protoconch comprises 1.3 whorls, measuring 440 µm in diameter, with slightly inflated initial part; nucleus measures 150 µm in diameter; entire protoconch surface weakly granulate; indistinct spiral striae appear on second half; P/T transition distinct, formed by sharp, thin axial line. Teleoconch whorls increase slowly but regularly in height and width; whorls moderately convex, whereas convexity slightly decreases with ontogeny. Last whorl attains 45-48% of shell height, passes over perfect convexity into slightly convex base. Aperture ovoid, inclined, closely attached to base of preceding whorl across almost entire parietal margin. Peristome thin, not thickened, little expanded; weakly sigmoidal in lateral view, with broad, shallow indentation in upper half and broad, weak protru-sion in lower half; inner lip protrudes in lateral view, extending sheet-like over base of penultimate whorl; umbilicus very narrow, slit-like. Growth lines weakly sigmoidal: strongly prosocline in upper half, weakly opisthocline in lower half. Several specimens show faint spiral threads on last and penultimate whorls.
Discussion. Our material matches well to the description of Issel (1865), corresponding in the conical shell shape, the regularly increasing whorls, the rounded last whorl with faint spiral striae, and the ovate, adapically angulated aperture; only his specimens (13.5 × 5 mm) are larger than ours and consist of more whorls. Compared to his description, Issel's illustration seem to overemphasize the relative height of the last whorl and underrepresent the pronounced whorl convexity. However, variability as to these characteristics is discernible also in our material.
Micromelania subulata Westerlund, 1902 is commonly listed as junior synonym of this species but always without discussion (e.g., Logvinenko and Starobogatov 1969, Kantor and Sysoev 2006, Vinarski and Kantor 2016. Westerlund's (1902b) description refers to a large (15 mm), elongate shell with 9.5-10 whorls and a thickened callus connecting the peristome margins. These features partly oppose Issel's description, which is why we tend to consider both taxa as separate, in contrast to most previous authors. Unfortunately, Westerlund's (1902b) type material of this species could not be traced, neither in the Göteborg Natural History Museum nor the Swedish Museum of Natural History in Stockholm, where the largest part of Westerlund's material is stored (Vinarski et al. 2013).
Another commonly cited synonym is Micromelania caspia var. inflata B. , which indeed matches both Issel's description and our material.
Turricaspia meneghiniana differs from the similarly large Laevicaspia caspia (Eichwald, 1838) in its regularly conical profile, the higher number of whorls, and the higher whorl convexity. The drawings of "Pyrgula meneghiniana (Issel)" provided by Logvinenko and Starobogatov (1969) indicate a broader shell with low whorl convexity and might represent a different species. In contrast, Pyrgula caspia sensu Logvinenko and Starobogatov (1969) (non Eichwald 1838) resembles the present species in terms of the high shell convexity and regular growth rate and might be conspecific. Turricaspia meneghiniana sensu Alexenko and Starobogatov (1987), with few, low convex whorls and an angled base, is clearly a different species.
Type material. Lectotype: ZIN 4422/1 (specimen illustrated by B. Dybowski andGrochmalicki 1915, 1917), designated by Logvinenko and Starobogatov (1969), illustrated by Kantor and Sysoev (2006: pl. 46 Description. Slender conical shell with up to 8 whorls. Protoconch bulbous, weakly granulate, with striae on second half; diameter 410 µm, consists of 1.25 whorls; nucleus low, broad, 140 µm wide; transition to teleoconch distinct. Teleoconch whorls weakly convex, with maximum convexity at or slightly below midline of whorl profile; portion above maximum convexity almost straight-sided, portion below weakly convex. Whorls are separated by deep suture. Height of last whorl amounts 45% of total shell. Sometimes intentions of spiral lines appear on lower half of last whorl. Aperture ovoid, oblique, with weakly thickened and slightly expanded peristome; in lateral view, peristome is distinctly sigmoidal, with broad, shallow indentation in upper half and broad, weak protrusion in lower half. Umbilicus very narrow or closed. Growth lines sigmoidal, markedly prosocline in upper half, weakly opisthocline in lower half. Discussion. The species can be easily distinguished from most other species of Turricaspia by its comparably broad conical shape, the low-convex whorls, and its small size. Juvenile specimens of T. meneghiniana remind of T. pulla but the former have broader shells with more convex whorls. Turricaspia pullula is likewise broader and exposes a characteristic tripartite whorl profile (see below).
Type locality. Caspian Sea (no locality specified). Dimensions. 5.36 × 2.62 mm. Description. A single incomplete specimen of about 6 whorls is preserved. Protoconch is corroded beyond recognition. Early teleoconch whorls are poorly convex to centrally flattened. Convexity strongly increases on about 3 rd whorl. From 4 th whorl onwards, whorl surface is partitioned into three zones: two lower zones are roughly straight-sided in profile, upper one slightly concave; middle zone slightly wider than other two; zones are separated by blunt angulations, whose expression varies between very faint to distinct (but no keel is formed). Aperture not preserved, but the tight coiling of the last preserved whorl suggests that umbilicus is absent. Growth lines strongly prosocline in upper third, near orthocline in lower two-thirds; transition coincides with boundary between upper and middle zone.
Discussion. The available specimen corresponds well to the lectotype as illustrated by Kantor and Sysoev (2006). The very characteristic tripartite whorl profile is only discernible on the penultimate whorl of their specimen. Such a pattern is unknown for any other Pontocaspian Pyrgulinae.
Description. Slender elongate shell, with up to nine convex whorls. Protoconch forms small bulbous cap, consisting of 1.3 whorls that measure 365 µm in diameter; surface weakly granulate, spiral striae set in after 0.5 whorls; nucleus is 140 µm wide; P/T boundary marked by thin, sharp axial line. Early teleoconch whorls have low convex profile. Two morphotypes are present: form A is broader, with whorls increasing slightly more in height (thus producing relatively larger last whorl) and little con- vex whorls; form B is more slender, whorls increase less fast in height in relation to width and whorl profile is stronger and more regularly convex. Both types are linked via intermediates. Generally, whorl profile varies between regularly convex (of varying strength), laterally flattened or bipartite (with near straight-sided upper half and convex lower half; rarely, transition between halves coincides with spiral thread). Suture is narrow. In some specimens, last whorl is slightly inflated and aperture is expanded. Traces of spiral sculpture, ranging from faint lines to blunt keels of variable number occur on several shells. Aperture expansion and sculpture are found on both morphotypes, as well as in intermediates. Umbilicus mostly covered by inner lip; if open, it is very narrow. Growth lines markedly sigmoidal, with prosocline upper third and opisthocline lower two-thirds.
Discussion. The huge morphological variability with intergrading morphotypes complicates reasonable taxonomic distinctions within this taxon. Moreover, much of the shape variation (especially in later whorls) seems to be a result of shell repair after predator-induced damage.
The variability also hampers linking our material to an existing name. Several species (and varieties) have been introduced for slender elongate, multi-whorled shells from the Caspian Sea. While the sculptured representatives can be fairly well delimited, the smooth-shelled taxa have caused considerable confusion. Particularly challenging are the many small, slender species with pointy apex, moderately to strongly convex whorls and thin peristome. The group includes (aside from T. spica): T. elegantula (Clessin & W. Dybowski in W. Dybowski, 1887), T. turricula , T. nossovi (Kolesnikov, 1947), T. concinna (Logvinenko & Starobogatov, 1969), T. spasskii (Logvinenko & Starobogatov, 1969), T. uralensis (Logvinenko & Starobogatov, 1969) and T. astrachanica (Pirogov, 1971). Turricaspia lyrata (B. , which was originally introduced as subspecies of T. spica, can be well delimited from that group because of its much larger, blunt apex. A major problem in identifying and discriminating those species is that the concepts applied by later authors occasionally diverge largely from the original perceptions. This especially regards T. spica and the species described by B. Dybowski and Grochmalicki (1915). Unfortunately, the types for these species are not known for sure (Kantor and Sysoev 2006, Vinarski andKantor 2016) and the original descriptions, drawings, and illustrations are mostly insufficient to allow distinction. Beyond that, different traits have been considered as diagnostic by different authors when describing new species, and morphological variability was hardly considered at all.
The identity of Turricaspia spica (sensu Eichwald) is dubious. The original description and illustration do not allow distinction from other similar species. The present specimens differ slightly from T. spica sensu Kantor & Sysoev, 2006, which is characterized by a faster whorl accretion rate and relatively higher whorls (including the last whorl). In contrast, our material largely fits the concept of T. spica as used by B. Dybowski and Grochmalicki (1917). We tentatively classify the Selitrennoye specimens in Turricaspia spica, being the oldest available name of the group. Many of the later proposed names might turn out to be junior synonyms. A more in-depth study is required to solve this problematic case.
Distribution. Turricaspia spica is endemic to the Caspian Sea. After Logvinenko and Starobogatov (1969), it occurs at a water depth between 0 and 30 m, but those authors applied a different concept of the species.
Discussion. The genus Abeskunus and the species that have been attributed to it have caused considerable confusion. A detailed discussion of the taxonomic and nomenclatural problems associated with Abeskunus, considerations on its systematic placement, as well as a description of the type species will be provided in a forthcoming study. Preliminary work confirms classification of the species described below in Abeskunus. ( Fig. 16B, C, I); 3.00 × 2.74 mm (RGM 1309842, Fig. 16D, E, G, H); 4.14 × 3.42 mm; 4.15 × 3.53 mm; 4.34 × 3.79 mm; 4.39 × 3.87 mm; 4.59 × 3.68 mm.

Abeskunus brusinianus
Description. Shell broadly conical, comprising up to 4.5 whorls. Protoconch broadly domical, with almost immersed initial part; consists of 1.25 whorls, measures 525 µm in diameter; nucleus is ca. 160 µm wide; nucleus surface covered with irregular elongated wrinkles; protoconch surface wrinkled, bearing thin, irregular spiral grooves on first third, passing over irregular pattern of faint spiral grooves and wrinkles into numerous, regularly parallel spiral furrows on last third; P/T transition without growth rim, marked by onset of growth lines. Teleoconch whorls highly convex, with maximum convexity in adapical half, producing slightly stepped spire. Last whorl attains 77-85% of shell height. Aperture drop-shaped, slightly inclined, with marked adapical notch at contact to penultimate whorl. Outer peristome margin not or slightly thickened, columellar and parietal margins weakly thickened; peristome slightly expanded towards columella (pro- truding towards umbilicus in lateral view) and strongly towards base; weakly sinuate in lateral view, with broad but weak adapical protrusion and straight-sided abapical part. Umbilicus narrow, slit-like. Prosocline growth lines cover shell surface.
Discussion. The species differs from the type species A. exiguus (Eichwald, 1838) in the conical shell, the taller spire, the less inflated last whorl, and the distinct umbilicus. Abeskunus brusinianus michelae Tadjalli-Pour, 1977 is much more globular than A. brusinianus. The latter species strongly reminds of and might be conspecific with A. exiguus. Pseudamnicola depressispira Logvinenko & Starobogatov, 1969, which these authors also included in the subgenus Abeskunus, differs from the presumed congeners in the valvatoid shape with very wide umbilicus and small but distinct riblets.
Distribution. Endemic to the Caspian Sea, in the southern and middle part at a depth of >250 m (Logvinenko andStarobogatov 1969, Parr et al. 2007).

Non-indigenous species
In addition to the Pontocaspian elements, six taxa including ubiquitous Palearctic species were identified. They all occur in low numbers and differ from Pontocaspian species in their preservation state. Shells of larger taxa (Valvata, Esperiana, and Lithoglyphus) are eroded and suggest transport. The smaller planorbids are better preserved but differ in their orange color indicating that they were not deposited along with the whitish shells of the Pontocaspian residents. Moreover, all six taxa are typical freshwater dwellers (e.g., Welter-Schultes 2012). They probably derive from rivers flowing into the northern Caspian Sea (Fig. 1).
Anisus cf. spirorbis (Linnaeus, 1758) (Fig. 17A-F). -13 juvenile to semi-adult specimens have been found. They match A. spirorbis as depicted by Glöer (2002) and Welter-Schultes (2012) in size, the regularly striated surface, the slightly overlapping whorls, and the weakly asymmetrical lateral profile. However, whorls expand in relative width a bit more rapidly in the Selitrennoye specimens, which is why we only tentatively assign our material to this species.
Planorbis cf. planorbis (Linnaeus, 1758) ( Fig. 17G-L). -Two juvenile specimens and one semi-adult are available, showing either a distinct or intentions of a keel on the periphery of the apical side, a feature typical of P. planorbis (Glöer 2002, Welter-Schultes 2012. In addition, shell size, whorl expansion and lateral profile fit well to this species. Since no adult specimen with fully developed keel have been found, we attribute our specimens to this species provisionally.
Bithynia sp. (Fig. 17M). -Four juvenile specimens are available, consisting of the protoconch and about one teleoconch whorl; the operculum is in-situ preserved in all specimens. The classification is based on the presence and shape of the operculum, the characteristics of the protoconch, as well as the shape of the aperture, all of which are features typical of the genus Bithynia (compare Glöer et al. 2005, Neubauer et al. 2016b. Esperiana esperi (Férussac, 1823) (Fig. 17N). -The single fragmentary specimen matches well the specimens illustrated by Welter-Schultes (2012). The similar and often co-occurring Microcolpia daudebartii acicularis (Férussac, 1823) is more elongate and lacks the color pattern (Glöer 2002, Welter-Schultes 2012. In the Russian literature, E. esperi is commonly listed as member of the genus Fagotia Bourguignat, 1884(e.g., Starobogatov et al. 2004, Vinarski and Kantor 2016, which is, however, invalid as a junior objective synonym of Esperiana. Starobogatov et al. (1992Starobogatov et al. ( , 2004 listed numerous species of Fagotia for extant European water bodies and categorized them into several subgenera. All of them are presently considered junior synonyms of Esperiana and E. esperi, respectively (for a complete synonymy list, see Vinarski and Kantor, 2016). Fagotia roseni Starobogatov in Starobogatov et al., 1992 from Quaternary deposits of Georgia also ranges within the variability of E. esperi and is herewith considered synonymous.
Valvata piscinalis (Müller, 1774) (Fig. 17P). -The eight, partly fragmented and corroded shells correspond well to Recent representatives of the species (Glöer 2002, Welter-Schultes 2012. Several of the Valvata species listed for the Volga delta region by Starobogatov et al. (1994, fig. 2) might be synonymous with this species. A conclusion on that matter requires examination of the material, which is unavailable to us.

Discussion
The current work provides a first insight into the magnitude of endemic Caspian gastropod biodiversity. The gastropod fauna of Selitrennoye is composed of 24 species, 16 of which are Pontocaspian endemic species and 15 exclusively Caspian. Six species are considered to be non-indigenous based on the combination of a truly freshwater autecology, a general wide spread palearctic distribution and a slightly different preservation from the bulk of the well preserved Caspian lacustrine species in the material. The non-Caspian gastropods are low in numbers, and we suspect they may have either floated into the Caspian Sea during periods of high river discharge or, more likely, were mixed in from underlying sediment layers through bioturbation. The Selitrennoye fauna was deposited in open lacustrine settings at a paleosalinity of approximately 10-11 psu as suggested by the general composition of the mollusk fauna (Yanina 2012). The presence of paired bivalves in the same sample indicates the in-situ character of the fauna. The shelly levels are located around 17 m b.s.l., and late Khazarian maximum sea levels are estimated as 10 m b.s.l. The presence of very sandy sediments with lenses suggests deposition above storm wave base. Altogether, this might translate into a sea floor at about 7 m water depth. The settings can be best compared with the present-day southermost part of the northern Caspian Basin.
The taxonomy and systematics of Caspian gastropods is very much in need of an update. The abundant and well-preserved material presented here has given an indication about the generic placements of species and the magnitude of species richness. When compared to the latest inventory of Caspian gastropods by Vinarski and Kantor (2016), who presented 92 species for the entire Caspian Sea, our numbers (that represent a single locality) are still rather low. The synonymization of species we propose points in general to lower species numbers for Caspian gastropod faunas as reported before. However, the possibility exists that some of the species considered synonyms are sibling species. In order to test for that, we will require extensive new living material to perform combined genetic and morphometric analyses. In recent expeditions in the coastal areas of Azerbaijan and in the Caspian territory of Kazakhstan, we failed to detect living endemic Pyrgulinae gastropod species. All Caspian endemics are suffering badly from invasive species that have caused a total turnover of the fauna during the 20 th century (Kosarev and Yablonskaya 1995, Grigorovich et al. 2003, Orlova et al. 2005, Therriault et al. 2004, Riedel et al. 2006, Heiler et al. 2010