Type genus: Leviathania Pčelincev, 1927.

Diagnosis.—Large sized shells with gradate spire and globular last whorl with wide sutural ramp. Sculpture consisting of nodes, which are confined to the shoulder. In many species, the nodes are only present on the spire whorls, if developed at all, whilst a bulge-like keel appears on the last whorl. Sculpture may also be completely reduced to a keel-like shoulder. No axial sculpture known so far. Umbilicus phaneromphalous. The inner wall of the umbilicus has a strong concavity reflecting a central sw elling of the inner lip; a weak anterior canal is present. Protoconch unknown.

Remarks.—The species attributed to Leviathania and its monogeneric family have usually been classified within the Purpurinidae Zittel, 1895 (e.g., Wenz 1939) or Purpuroideidae Guzhov, 2004. The main argument for this assignment seems to have been the large last whorl and the angulated whorls. Kaim (2004) provided a revised diagnosis of the type genus of the Purpurinidae, which characterises members of this family as moderately sized with prominent axial and spiral sculpture. The umbilicus is reduced to a chink, which is covered by a thin inner lip. Numerous nominal species of Purpurina d'Orbigny, 1850 have been described by Eichwald (1868), Hudleston (1888), Brösamlen (1909), and Kaim (2004). The morphology of these taxa matches the diagnosis of Purpurina by Kaim (2004) and clearly excludes a close relation with the Leviathaniidae fam. nov. A revision of the family Purpurinidae Zittel, 1895 was presented by Guzhov (2004) who includes the genera Angularia Koken, 1892, Eucycloidea Hudleston, 1888, Moerckeia Böhm, 1895, Pseudoscalites Kittl, 1894, Purpurina d'Orbigny, 1850, Tretospira Koken, 1892, and Ptychostoma Laube, 1868. Following earlier suggestions by Golovinova and Korotkov (1986), Guzhov (2004) doubted that Purpuroidea Lycett, 1848 and Leviathania Pčelintzev, 1927 can be maintained within that family and consequently introduced the new family Purpuroideidae for these genera. He based this family on “large or extremely large multispiral shells that lack an umbilicus and possess a siphonostomatous aperture bearing a short half-closed siphonal canal.” This diagnosis fits well to the numerous representatives of Purpuroidea Lycett, 1848, as listed by Guzhov (2004), but obviously excludes Leviathania, which possesses a very distinct and deep umbilicus. Already Guzhov (2004) discussed the possibility that Leviathania should be removed from the Purpuroideidae but refrained from doing so due to the inadequate material known to him.

In the original description, Pčelincev (1927) considered this genus to be anomphalous. This misinterpretation might be explained by the fact that the gastropods are represented only by internal moulds consisting of the identical sediment as the umbilical filling. Therefore, the umbilical area differs hardly from the rest of the gastropod and appears to be closed at first sight. The character of the umbilicus is usually revealed only if cross sections are studied, allowing to trace the otherwise enigmatic shell outline.

Type speciesleviathan: Natica leviathan Pictet and Campiche, 1863 (the junior synonym of Strombus sautieri Coquand, 1856). Sainte-Croix (Kanton Waadt, Switzerland), “calcaire roux dans les marnes à bryozoaires” (Pictet and Campiche 1863), Valanginian.

Species included: Leviathania gigantea (Makowsky, 1874), Tithonian, Czech Republic and Austria; Leviathania sp., Tithonian or Berriasian, Ukraine; Leviathania sautieri (Coquand, 1856), Valanginian, Switzerland, France; Leviathania gerassimovi Pčelincev, 1927, Valanginian, Caucasus; Leviathania subnodosa (Roemer, 1836), Kimmeridgian, Northern Germany (Fig. 2). Doubtful species which probably belong to Leviathania, but are based on poorly preserved specimens: Leviathania sulcata Pčelincev, 1963, Cretaceous, Crimea; L. sp., Tithonian, Crimea (Korotkov 1989); L. beschterikensis Golovinova, 1982, Cretaceous, Crimea; L. gabonensis (Meunier, 1887), Albian?, Gabon.

Diagnosis.—As given for family Leviathaniidae. If Tuberleviathania Golovinova and Korotkov, 1986 might be accepted as valid genus, the definition of Leviathania could be reduced to species with weak to obsolete nodes on the last whorls.

Discussion.—The species detailed below was usually placed in the genus Purpuroidea Lycett, 1848. Purpuroidea is still used in a much too broad sense and needs revision as done by Kollmann (1984) for alleged Late Cretaceous representatives of that genus. In fact, this genus is based on the type species Murex? nodulatus Young and Bird, 1828 from the Bathonian of England (note that Wenz [1939] erroneously mentioned Purpuroidea maureausea [Buvignier, 1852] as type species). This species has a steep sutural ramp, blunt nodes and a rugose axial sculpture on the last whorl and develops a basal spiral cord. Its spire is high and coalescing. Other representatives of Purpuroidea, as described by Buvignier (1852) and Hudleston (1880), are also characterised by a moderately high spire and strong sculpture of the last whorl, which is bulbous naticoid. These morphologies have little in common with the huge shells from the Tithonian of Austria and the related specimens from Ukraine. In contrast, Leviathania Pčelincev, 1927 comprises large species with a gradate spire, smooth shell and anteriorly strongly expanding aperture. The sculpture is restricted to nodes or a keel along the shoulder. Its Valanginian type species is Natica leviathan Pictet and Campiche, 1863, which is a junior synonym of Strombus sautieri Coquand, 1856. Thus, it has to be referred to as Leviathania sautieri (Coquand, 1856).

Several species which have been assigned to Leviathania by Pčelincev (1927) and Golovinova (1982) from the Cretaceous of Russia have been transferred to Tuberleviathania Golovinova and Korotkov, 1986, with the type species Leviathania gerassimovi Pčelincev, 1927. Golovinova and Korotkov (1986) separated this genus from Leviathania based on the presence of nodes on the shoulder, the large size and the broad last whorl. The size and the width of the last whorl, however, are no distinguishing features. The size is even topped by some Leviathania species, as discussed below, and the broad last whorl is also developed in the type species of Leviathania, L. sautieri. Only the prominent nodes would be of interest, but there are also Tuberleviathania species with weakly developed nodes (Guzhov 2004). Moreover, all taxa which have been assigned to Tuberleviathania in the literature are based on poorly preserved internal moulds and are rather dubious. Therefore, we follow Guzhov (2004) and treat Tuberleviathania Golovinova and Korotkov, 1986 tentatively as synonym of Leviathania.

Stratigraphic and geographic range.—Kimmeridgian to Valanginian: Portugal, Spain, France, Switzerland, Austria, Czech Republic, Germany, Ukraine, Crimea, Caucasus, Gabon (Green 1934; Foucault 1971; Bosca 1922; Pictet and Campiche 1863; Makowsky 1874; Alth 1881, 1882; Pčelincev 1963, Golovinova 1982; Golovinova and Korotkov 1986; Meunier 1887).

1874 Pterocera gigantea nova species; Makowsky 1874: 123, pl. 2 [non Rostellaria gigantea Geinitz, 1842].

1910 Purpuroidea siibnodosa, Abel 1910: 9.

1940 Purpuroidea siibnodosa Roemer; Bachmayer 1940: 14.

1954 Purpuroidea siibnodosa Roemer; Bachmayer 1954: 462, 463, fig. 20.

1964 Purpuroidea siibnodosa (Roemer); Bachmayer 1964: 104; fig. 124.

1969 Purpuroidea siibnodosa (Roemer); Bachmayer 1969: 116; fig. 141.

1974 Purpuroidea siibnodosa Roemer; Thenius 1974: 58, fig. 112/20.

2002 Purpuroidea siibnodosa; Plöchinger and Karanitsch 2002: 42, 198, figs. 33, 431.

2006 Purpuroidea; Wessely 2006: 74, fig. 148.

Type material: Neotype: specimen NHMW2011/0347/0001, height: 268 mm, width: 225 mm (Figs. 3A, 4B).

Type locality: Dörfles “Quarry 5” sensu Bachmayer (1940), GPS WGS 84: N48°32′37.28”, E16°20′58.12”, north of Ernstbrunn, ca. 50 km north of Vienna; Lower Austria.

Type horizon: Ernstbrunn Limestone (Ernstbrunn formation); Middle to Late Tithonian (Zeiss and Hofmann 2001).

Material.—Reference specimen 1: NHM 2011/0264/0001: height: 255 mm, width: 195 mm (Fig. 3B). Reference specimen 2: NHMW2011/0347/0002, height: 182 mm, width: 158 mm (Fig. 3C).

Diagnosis.—An extraordinary large species, attaining 300 mm in height, with gradate spire and nearly flat sutural ramp. Early whorls bear indistinct shoulder nodes, which are replaced by a bulgy keel on later whorls. The bulbous last whorl is very large, rapidly expanding in diameter compared to spire, with a distinct concavity below the keel. A narrow “columellar” fold occurs close to the suture on the columellar wall, which develops a very prominent central swelling. Aperture anteriorly widening; phaneromphalous umbilicus deep and wide with dense sculpture of scaly growth increments.

Description.—Very large gastropods, which have attained up to 30 cm in height. The protoconch and the earliest teleoconch are unknown. Teleoconch consisting of at least 7–8 whorls. The earliest preserved whorls are moderately convex with a faint median angulation; no sculpture is detectable on the available material at that stage of growth. The angulation shifts quickly into the upper portion of the following whorl and grades into a very prominent shoulder. This angulation separates a feebly concave sutural ramp from the nearly straight sided to slightly convex outer face which is subparallel to the spire axis. The suture is thread-like without incision and adjoined by a delicate, thin sutural collar. The slight concavity of the ramp is caused by the indistinct, wide spaced knobby swellings at the angulation, which are directed in adapical direction but do not protrude abaxially over the periphery. These swellings become blunt nodes on the subsequent spire whorl. Although the exact number cannot be counted, they may have ranged around 15 on the penultimate spire whorl. There, the nodes become spirally elongate, low and poorly defined, but still with adapical orientation. On the last spire whorl and the body whorl, the nodes disappear completely and are replaced by a prominent bulgy keel. This structure pretends a somewhat allometric growth with quickly broadening sutural ramp. The ramp is divided into a weakly convex adsutural part and a shallow concavity close to the bulge. Similarly, the outer face develops a considerable concavity below the bulge, which is a very characteristic feature of this species. On the last whorl, this concavity passes into the strongly convex but rapidly contracting base. The peristome is lost in all specimens, but the preserved terminal parts of the last whorl show that the aperture widens significantly anteriorly. Consequently, the base forms a marked neck. The entire shell is smooth except for faint growth lines. On spire whorls, these comprise a strongly prosocline part along the outer face which passes via a shallow sulcus into an opisthocline adsutural part on the ramp. On the broad sutural ramp of the last spire whorl, the growth lines are sigmoidal with a shorter prosocyrt part close to the bulge and a wider, deeper opisthocyrt adsutural part.

The columella is divided into a deeply concave anterior part which passes into a prominent convex central swelling, followed by a slightly narrower posterior concavity. The surface is smooth except for a narrow but strongly raised fold situated a few millimetres below the suture. Due to the central swelling, the inner lip is strongly curved, robust and becomes adapically thinner, forming a thin columellar plate at the base. Anteriorly, it thickens considerably and passes into a broad and blunt fasciole. This structure borders a phaneromphalous, perspectivic umbilicus which reaches up to the earliest spire whorls. A strong concavity in the umbilicus causes the prominent central swelling in the columellar wall (Fig. 4A₂, A₃, B). The most characteristic feature of the umbilical surface is a dense succession of growth increments, which cause a rough and scaly sculpture (Figs. 4A₁, A₂, 6). A weak anterior canal is developed. Its exact morphology and transition into the outer lip is unknown. The prosocline growth lines of the outer face seem to define also the margin of the outer lip (Fig. 3B₁).

Remarks.—Makowsky (1874) correctly recognized that this huge species is still undescribed and proposed Pterocera gigantea as new name. He was not aware that this name was preoccupied by the Cretaceous Rostellaria gigantea Geinitz, 1842, which has been placed into Pterocera by Reuss (1846), and thus created a junior secondary homonym. However, since Pterocera gigantea Makowsky, 1874 is now referred to Leviathania, the junior secondary homonym becomes available (ICZN 1999: Article 59.2), and the valid species name is Leviathania gigantea (Makowsky, 1874).

The holotype of Pterocera gigantea Makowsky, 1874, originally part of the collections of the Naturforschender Verein Brünn, has been lost (personal communication Růžena Gregorová, Moravian Museum, Brno). Makowsky (1874) stated that his specimen had been collected from a quarry near Bergen (nowadays Perná) north of Mikulov. Since no outcrops of Ernstbrunn Limestone exist in the surroundings of Perná, the fossil gastropod was presumably derived from the southern part of the Děvin hill. As already stated by Makowsky (1874), no additional fossils have been reported from this outcrop. In order to clarify the taxonomic status of the species, we herein select a fully grown, well-preserved specimen of Leviathania gigantea from the Ernstbrunn Limestone of Dörfles near Ernstbrunn (Lower Austria) as a neotype.

Most authors, aside from Makowsky (1874), have identified the specimens as Purpuroidea subnodosa (Roemer, 1836). This species derives from the Kimmeridgian Süntel Formation of Northern Germany (Fig. 5A; Menning and Hendrich 2002) and is thus distinctly older than the Tithonian Leviathania gigantea. Aside from the stratigraphic gap, it differs considerably from L. gigantea in its smaller size, the steeper sutural ramp and the absence of a marked shoulder-bulge. It develops also a concavity on the inner wall of the umbilicus, but this is w eaker compared to L. gigantea. The type of this species is lost, but material from the Roemer collection in the Roemer- und Pelizaeus-Museum Hildesheim and specimens from the NHMW suggest that L. subnodosa developed prominent and large nodes even on the last whorl. The general shape and the deep and wide umbilicus place this taxon clearly within the newly established family Leviathaniidae.

Leviathania sautieri (Coquand, 1856) (= L. leviathan [Pictet and Campiche, 1863]), from the Valanginian of France and Switzerland, represents a closely related species (Fig. 5C). It agrees largely in outline with the characteristic gradate spire and broad sutural ramp on the last whorl. The spire, however, is much lower and the sutural shelf of the last whorl is wider. It develops a keel, which is slightly less bulgy but lacks any nodes on the spire whorls. It is only slightly smaller and seems to lack the anterior widening of the aperture (cf. Pictet and Campiche 1863: 562, pl. 89: 1, 2). The concavity on the inner wall of the umbilicus of Leviathania sautieri is distinct (Fig. 5C) but less prominent.

Alth (1881: 29, pl. 2: 1; 1882: 207, pl. 19: 1) described a specimen from the Ukrainian Nyzhniv Formation as Purpurina subnodosa, which is clearly related with Leviathania gigantea. This gigantic specimen attained nearly 40 cm in height. It differs from L. gigantea in the absence of the characteristic concavity below the shoulder and lacks a distinct bulge. Further, it lacks nodes on early spire whorls and the nearly straight outer face of the spire whorls is slightly scalate. Finally, its spire is higher. These differences suggest that the Ukrainian shell is not conspecific with Leviathania gigantea, which displays a quite constant morphology. Except for the outstanding size, the Ukrainian species is intermediate between the Tithonian L. gigantea and the Valanginian L. sautieri (Coquand, 1856). We consider it to represent a distinct, yet unnamed, species, which already heralds the typical Cretaceous morphology. Therefore, we did not include this reference in the chresonymy of Leviathania gigantea. Since we have not studied any specimens from Nyzhniv, we refrain from assigning a new name.

Leviathania sulcata Pčelincev, 1963, from the Cretaceous of Crimea, is stout and has a convex shoulder without nodes or a bulge but is reminiscent of our species concerning the widening of the aperture (cf. Pčelincev 1963: 49, pl. 15: 1). Leviathania beschterikensis Golovinova, 1982 differs from Leviathania gigantea in its strong nodes on the shoulder and the more regular growth, resulting in a comparatively slender shell. Leviathania gerassimovi Pčelincev, 1927 from the Valanginian of the Caucasus differs in its convex whorls and the rapidly increasing diameter of the whorls; it lacks a sutural ramp.

The stratigraphically closest species with similar morphology is Purpuroidea carpathica Zittel, 1873 (Zittel 1873: 316, pl. 343: 3, 4) from the Tithonian of Štramberk. This species differs from the new species in its distinctly smaller size of less than 15 cm height, the prominent nodes and the higher spire. The growth lines are steeply prosocline on the sutural ramp, thus differing strongly from L. gigantea.

Distribution.—Leviathania gigantea is frequently found in the Tithonian Ernstbrunn Limestone in the quarries at Dörfles and Ernstbrunn in Lower Austria and Mikulov in southern Moravia (Czech Republic). Additional Tithonian material was found at Štramberk in Czech Republic (private collection of Lubomir Martinásek, Kopřivnice, Czech Republic).

Discussion.—The whitish to light-grey Ernstbrunn Lime-stone typically consists of almost 100% of calcium carbonate (Moshammer and Lobitzer 1997). At Dörfles Quarry 5 (denomination sensu Bachmayer 1940), which has yielded most of the discussed gastropods, the rocks comprise a patchwork of slightly variant, highly fossiliferous facies, dominated by bio-pelmicrites (wacke- and packstones), bio-pelsparites, algal bindstones and oncoid pack- and grainstones. Characteristic facies fossils include dasycladalean, codiacean, and solenoporacean algae, foraminifers, nerineid gastropods, and epidiceratid, lucinid and cardioid bivalves. Small scleractinian and coralline sponge patch reefs occur locally (Hofmann 1990).

Bachmayer (1957) proposed a diversity of more than 500 fossil species for the Ernstbrunn Limestone, which evidently was no overestimate. With regard to species numbers, the assemblage is macroscopically dominated by gastropods, bivalves, and crustaceans. Still relatively diverse are ammonites, scleractinian corals, echinoderms, and brachiopods. Among these groups, the bivalves presumably accounted for the major portion of biomass, since they were represented by numerous extraordinarily large species of, e.g., Epidiceras, Pachyrismella, and Pterocardia, and abundant Astartidae and Lucinidae. Gastropods also comprise several large taxa, mainly within the Pleurotomariidae, Aporrhaidae, and Nerineidae. The latter occur in large quantities, and represent one of two major components of the epidiceratid-nerineid biofacies distinguished by Hofmann (1990). In this respect, dasycladalean algae also have to be considered, as they dominate the assemblage at places, and surely played an effective role in the production of biomass. The largest fossils of the Ernstbrunn Limestone community, however, are the giant Leviathania gastropods.

A similar assemblage is documented from the Nyzhniv Formation in Ukraine by Alth (1881, 1882). He described a total of 180 species, 93 of them gastropods, and 57 bivalves. In contrast to Ernstbrunn-Dörfles, epidiceratids are only represented by a few extremely small specimens, and thus not important for biofacies. Crustaceans are represented only as indeterminable fragments. Interestingly, four species of Dasycladales are also present.

Although a palaeoecological evaluation of the respective assemblages is lacking, a comparable environment can be inferred for Leviathania sautieri. On one hand, the gastropod mould from Mont Salève is composed of bio-pelmicrite with small oncoids (Fig. 5C); on the other hand, the fauna from Sainte-Croix comprises many taxa typical of lagoonal settings (e.g., nerineids; Pictet and Campiche 1861–1864).

Hence, tropical lagoonal environments of the northern coasts of the Tethys Ocean seem to have been the preferred habitat of the Leviathaniidae. This is also supported by the comparatively scarce occurrences in the reef and fore reef settings represented by the coeval Štramberk Limestone.

The Leviathaniidae have no living relatives and the interpretation of their palaeoecology remains speculative. The large number of specimens in the collection of the NHMW is not a sampling bias due to the attractive size of these fossils, but indeed reflects the high abundance of the gastropods in the field, which can still be observed in the quarries around Ernstbrunn. Obviously, this species lived in huge populations. The shells are rarely associated with scleractinian and coralline sponge patch reefs but are typically filled with bio-pelmicrites, bio-pelsparites, and oncoid pack- and grain-stones (Fig. 5). Hence, the animals seem to have preferred the rather mobile bioclastic sediment which provided a sandy substratum. In modern marine faunas, only herbivore strombids are found in comparable size and population densities in shallow marine, lagoonal settings. The huge Lobatus gigas (Linnaeus, 1758) occurs in densities of 70–80 specimens/ ha in the Bahamas and the Dominican Republic (Stoner and Ray 1996; Posada et al. 1999). A direct comparison of the ecology of both taxa, however, is hampered by the fact that the extant strombid is largely bound to seagrass meadows, which have not been in existence in Jurassic times (Eva 1980; Hemminga and Duarte 2000). Another huge strombid, which forms considerable populations is Lambis truncata ([Lightfoot], 1786), which was counted to occur in densities of 1.3 specimens/ha in protected lagoons in Kenia (McClanahan 1989). Other extant gastropod species of comparable size are the carnivores Cassis cornuta (Linnaeus, 1758) and Charonia tritonis (Linnaeus, 1758), which occur in distinctly lower densities even in areas which have not been exposed to fishery (McClanahan 1989). Therefore, we suggest that the abundance of the large-sized Leviathania might rather correspond to a herbivorous/omnivorous life style comparable to the extant strombids.