The investigation of the Late Paleogene to Late Neogene species of rissoid gastropod Galeodinopsis in the Euro-Mediterranean area has supported the hypothesis that this genus is an intermediate form between two well-known rissoids, Alvinia and Manzonia. We recognized four species of Galeodinopsis: G. biangulata, G. germanica sp. nov., G. semperi (new name for Rissoa duboisii), and G. tiberiana. The oldest (very Late Eocene/Oligocene) representatives of Galeodinopsis, G. biangulata, and G. semperi, share similar shell shape and microsculpture with Alvinia. This suggests that Galeodinopsis originated from some Eocene species related to Alvinia. The new species represents the first occurrence within Galeodinopsis of a combination of characters very close to those of Manzonia, above all the typical pitted microsculpture. We hypothesize that Manzonia evolved from Galeodinopsis rather than from the genera Alvania or Alvinia, as previously supposed. Galeodinopsis originated during the very Late Eocene/Oligocene in the North Sea Basin. Afterwards it underwent a strong southward shift to the mid-high east Atlantic and the Mediterranean area, during the Mio-Pliocene, and to the Recent tropical eastern Atlantic coasts, where the type species G. tiberiana still lives. The shift likely was due to a combination of climate cooling and palaeogeographical changes. The distribution of G. biangulata suggests that connections between the North Sea Basin and the Atlantic domain opened through the Channel area at least during the Early Oligocene, earlier than indicated previously. The distribution of Manzonia moulinsi supports the idea of a southern connection to the Atlantic Aquitaine Basin via the Rhine Graben during the Late Oligocene. From a palaeoecological point of view, Galeodinopsis includes warm species with planktotrophic larval development that are typical of the shelf environment in fully marine conditions.
The family Rissoidae includes some of the most diverse gastropod groups, with numerous genera inhabiting a great variety of environments, from shallow to bathyal depth, in the Mediterranean, Atlantic, Indo-Pacific, and the temperate Australian and New Zealand coasts (Ponder 1985). The first record of this family dates back to the late Lower Jurassic of Europe, with a discrete differentiation in Middle Jurassic (Ponder 1985; Conti et al. 1992; Kaim 2004; Schulbert and Nützel 2009). Strong diversification occurred during the Tertiary, particularly in the Euro-Mediterranean area (Sacco 1895; Seguenza 1903; Cossmann 1921; Lozouet 1998; Kowalke and Harzhauser 2004; Landau et al. 2004; Chirli 2006).
As with other small gastropods, most fossil rissoid shells preserve the larval shell (protoconch), which provides information on the larval biology of their respective species (Shuto 1974; Jablonski and Lutz 1983) and, in several cases, provides a helpful taxonomic tool for separating taxa (Ponder 1985). However, because of the high degree of diversification and confusion caused by convergence in shell characters, most of these gastropods are difficult to classify, particularly at the genus and species level. As a consequence, several attempts to separate groups at genus-subgenus level, especially within the very polymorph genus Alvania Risso, 1826 and closely related nominal genera, failed or simply showed the differences to be inconsistent (Ponder 1985; Garilli 2008; Garilli and Parrinello 2010). Conversely, successful results have been achieved in the Late Paleocene—Late Neogene genus Galeodinopsis Sacco, 1895 on the basis of an accurate taxonomic approach involving a detailed morphological study of protoconch and teleoconch characters, with special regard to teleoconch microsculpture (Garilli 2008). Notwithstanding this, the actual number of species within Galeodinospis is not at all defined, mainly due to the lack of a complete taxonomic setting. Actually several taxa, such as Alvania, Alvinia Monterosato, 1884, Arsenia Monterosato, 1891, Flemellia Nordsieck, 1972, and Manzonia Brusina, 1870, have been or could be related to Galeodinopsis but without the support of taxonomical comparative analyses based on in depth studies of teleoconch and protoconch features involving type species. Consequently the origin of Galeodinospsis and its possible phyletic relationships with close genera are still to be defined.
The main aim of this study is to provide an exhaustive taxonomic data set of Galeodinopsis allowing a rediscussion of its phyletic link to the other closely related rissoid taxa. We therefore describe in detail protoconch and teleoconch features of Galeodinopsis species from the Tertiary of central Europe and the eastern Atlantic-Mediterranean area (Fig. 1). Comparative studies between Galeodinopsis and type species of related genera are also provided. The palaeoecological and paleogeographical occurrences of Galeodinopsis is discussed. A new species of this genus from the Cenozoic of north-central Europe is also described.
Institutional abbreviations.—BGR, Bundesanstalt für Geowissenschaften und Rohstoffe, Berlin, Germany; IRSNB, Institut Royal des Sciences Naturelles de Belgique, Brussels, Belgium; MGUP, Museo Geologico G.G. Gemmellaro, Università di Palermo, Italy; MNHN, Muséum National d'Histoire Naturelle, Paris, France; MNHU, Museum für Naturkunde der Humboldt Universität, Berlin, Germany; MPOB, Museo di Paleobiologia e dell'Orto Botanico, Università di Modena and Reggio Emilia, Modena, Italy; MZB, Museo di Zoologia dell'Università di Bologna, Italy; MZPD, Museo di Zoologia P. Doderlein, Università di Palermo, Italy; SMF, Senckenberg Forschungsinstitut und Naturmuseum, Malakologie, Frankfurt am Main, Germany; UCBL, Université Claude Bernard Lyon 1, Lyon, France.
Other abbreviations.—AR, Agatino Reitano collection (Tremestieri Etneo, Catania, Italy); Coll(s)., collection(s); Ht, total shell height; EM, collection Ecole Nationale Supérieure des Mines de Paris; MF, Maurizio Forli collection (Prato, Italy) ex Stefano Palazzi collection (Milo, Catania, Italy); VG, Vittorio Garilli collection; W, shell width.
Sacco (1895) described Galeodinopsis as a subgenus of Alvania, but Ponder (1985) did not follow this separation and considered Galeodinopsis to be a synonym of Alvania. Recently, Galeodinopsis has been re-evaluated as a distinct genus representing shells with characters shared by the genera Alvania and Manzonia Brusina, 1870 (Garilli 2008). In particular, mainly on the basis of the characters of the type species, the Miocene to Recent species Rissoa tiberiana Coppi, 1876 (Fig. 2), Galeodinopsis is characterized by an Alvania-like shape, with an inflated last whorl usually bearing varices and a large aperture resembling that of Galeodina Monterosato, 1884 (= Alvania according to Garilli 2008), and a Manzonia- like combination (see also below) of characters such as the double-rimmed outer lip, the teleoconch pitted microsculpture, and the netted protoconch 1 (Garilli 2008). Apart from the type species, only the Oligocene species “Rissoa Duboisii Nyst, 1845” has been included in Galeodinopsis, whereas the Recent Macaronesian species R. spreta Watson, 1873 was doubtfully considered a close species (Garilli 2008). The fossil species R. multicostata Speyer, 1864, compared with G. tiberiana by the same author, was tentatively placed in Alvinia, but without an in-depth critical review.
The genus Manzonia (Fig. 3), type species Turbo crassus Kanmacher in Adams, 1798, includes species most of which are characterized by ovate-conic-elongate shells bearing robust axial sculpture, formed by strongly curved to sinuous, opisthocline ribs in Manzonia s. s., and strong, smooth spiral cords on the base of the shell. Other relevant characters are the markedly duplicated peristome, the netted sculpture of protoconch 1 (in multispiral protoconchs assumed to indicate planktotrophic larval development; Fig. 3G2) and a characteristic teleoconch microsculpture, consisted of a pitted surface on flat spiral cords (Ponder 1985; Moolenbeek and Faber 1987a) and roughly prismatic elements forming numerous, very fine spiral threads in the interspaces between primary cords (Garilli 2008; see Fig. 3A2, B2, G4). Manzonia shells do not bear varices (apart from that surrounding the outer lip); to our knowledge the sole specimen with a varix on the last whorl is a Manzonia sp. from the early Late Miocene of Montegibbio (Modena, northern Italy) (Fig. 3H). Subgeneric division within Manzonia is still debated as well its degree of affinity to Alvania. The classification of Alvinia Monterosato, 1884 (type species Alvania weinkauffi Weinkauff, 1868 ex Schwartz von Mohrenstern ms) as a subgenus of Manzonia, as proposed by Ponder (1985) and accepted by Kowalke and Harzauser (2004), was not followed by Moolenbeek and Faber (1987c). Even the consideration of Flemellia Nordsieck, 1972 (new name for Seguenziella, Sacco, 1904, new name for Taramellia Seguenza, 1903, new name for Flemingia Jeffreys, 1884; see Ponder 1985) as a Manzonia-linked taxon appears to be controversial, because its type species, Turbo zetlandicus Montagu, 1815, was only provisionally placed in Manzonia (Flemellia) by Moolenbeek and Faber (1987c), while Ponder (1985) considered it to be a species of Manzonia (Alvinia), as he considered Flemellia to be a synonym of Alvinia. Conversely, Bouchet and Warén (1993) considered Flemellia to be a synonym of Alvania. The type species of Alvinia has a turrited-shouldered shell lacking the typical microsculpture and the marked duplication of the peristome (Figs. 4A, B, 5B) seen in the type species of Manzonia, and its protoconch is paucispiral and characterized by discontinuous spiral cords of rough tubercles. Alvania, type species Turbo cimex Linnaeus, 1758, and Arsenia, type Turbo punctura Montagu, 1803, have a conical-ovate shell without a shoulder, but with a simple outer lip; protoconch 1 in Alvania and Arsenia is characterized by spiral elements and very fine dots, and the microsculpture is weak to obsolete (Figs. 6, 7C–E). Some Alvania species bear stronger microsculpture with discontinuous fine threads representing a sort of Alvinia-impoverished pattern (Garilli 2008; Garilli and Parrinello 2010). The taxonomical state of the art is also complicated by the features of species such as T. zetlandicus and Rissoa testae Aradas and Maggiore, 1844 (type species of Actonia Monterosato, 1884, regarded as synonym of Alvania by Ponder  and Bouchet and Warén ): T. zetlandicus has a shell shape and microsculptural pattern (Figs. 4C1, C2, D, 5A) similar to those seen in the type species of Alvinia, and shares the netted pattern of protoconch 1 and the duplicated peristome (Fig. 4C1, C3, C4) with Manzonia s.s.; R. testae (Fig. 8) has shell shape and protoconch characters comparable to those of Manzonia, but has a microsculpture similar to that of Alvinia and Flemellia. All this makes the relationship between Alvania, Alvinia and Manzonia yet to be established, despite Gofas (2010), supporting Ponder's (1985) view, emphasizing that even species lacking the pitted microsculpture may be regarded as closer to Manzonia than to Alvania. Ponder (1985) also suggested that Manzonia (Alvinia) may have diverged from Alvania early in the Tertiary.
An exhaustive palaeontological dataset on these arguments, with emphasis on possible transitional steps between most of these taxa, has never been provided. The only well-established Manzonia species from the Paleogene of Europe are Alvania foraminata Lozouet, 1998, from the French Late Oligocene, Chattian, and Rissoa moulinsi d'Orbigny, 1852 known from the Chattian and Miocene of central-western Europe, both showing all the main distinctive characters mentioned above (see Lozouet 1998: fig. 9f–j; Garilli 2008). More species occurred during the European Neogene (Kowalke and Harzhauser 2004; Garilli 2008 and references therein), where diversification involved several indisputable Manzonia taxa such as Rissoa (M.) falunica Morgan, 1915, R. (M.) pontileviensis Morgan, 1915, Cyclostoma scalare Dubois, 1831, Flemingia subzetlandica Boettger, 1907 (possibly = M. zetlandica var. miocrassicosta Sacco, 1895; see Kowalke and Harzhauser 2004: fig. 8A; Fig. 3). Today, one of the most Manzonia species-rich provinces is represented by the northeastern Atlantic seamounts and the Macaronesian archipelago (Moolenbeek and Faber 1987a–c; Gofas 2007), where the high diversification is testified to by species somewhat morphologically distant from the type species, but showing some typical characters. Apart from doubtful Late Cretaceous records (Ponder 1985), the earliest occurrence of Alvania is from the Eocene of Europe (Le Renard 1990; MNHN Fossil Database of Tertiary shells and references therein), where this genus underwent strong diversification during Neogene—Recent time (e.g., Sacco 1895; Bouchet and Warén 1993; Giannuzzi-Savelli et al. 1996; Chirli 2006). As for Alvinia, less is known about the palaeontological record, although it is clear enough that species resembling the type species occurred earliest in the Eocene, Lutetian, of France (Le Renard 1990; MNHN Fossil Database of Tertiary shells, with references), also with shells not at all distant from Alvania. It assumed the characteristic turrited shape in the Miocene of the central Paratethys and of the North Sea Basin (see Bałuk 1975, as Alvania (Taramellia); Kowalke and Harzhauser 2004, with references).
Material and methods
The examined material is housed in several European institutions. Part of this material was provided by private collectors or collected by the first author, and is now housed in MZPD. Another part, used for comparison, comes from samples collected during the oceanographic field trip CS'96 in the Strait of Sicily (vessel “Urania”, Chief Scientist Marco Taviani) and is now housed in MZB.
In the parts dedicated to each species of Galeodinopsis, a list of the material is provided with all information given on the original labels. An abbreviated list of essential synonymy and/or chresonymy is also provided. For all the discussed species, four to five shells were studied by SEM, using LEO 420 SEM and Philips XL30 ESEM. SEM pictures of the type specimen of Rissoa duboisii from Nyst coll. were provided by IRSNB. Those of material at UCBL were provided by Arlette Armand (U.F.R. Sciences de la Terre, Lyon, France). With the exception of material loaned by public institutions, specimens examined by SEM were cleaned in a Bransonic 5 ultrasonic machine using distilled water. Specimens from public institutions were cleaned using a very fine brush wet with distilled water. Particular attention was given to the protoconch and teleoconch microsculpture as sources of taxonomic characters at species and genus level. The number of protoconch whorls was counted according to Verduin's (1977) method. Shells were measured using a stereomicroscope provided with a cross-line micrometer eyepiece and/or by SEM micrometer tool. The type specimen of Rissoa duboisii was measured on the SEM images. All the shells contained in each studied lot were measured with the exception of Galeodinopsis tiberiana from MF (with more than 200 shells), for which only selected specimens were measured, in particular those assumed to be at terminal growth. Dimensions of the shell at MNHU (catalogue number MB.Ga.43) are from Tembrock (1964: pl. 7: 6). For lots containing numerous shells, measurements are given for selected specimens together with average value and standard deviation. Measurements are shown in Supplementary Online Material at http://app.pan.pl/SOM/app59-Garilli_Parrinello_SOM.pdf. The position of varix/varices on the last whorl is indicated in degrees of the angle formed by the plane of the varix and the outer lip. Geographical and stratigraphical distributions of each discussed species are based on the examined material and the literature. Published records were critically evaluated on the basis of good illustrations, satisfactory descriptions and cross-checked references. As for material from institutions and private collections, in general we followed the stratigraphical attribution reported by the collectors.
Stratigraphical notes on some recovery areas
Agatino Reitano provided part of the analysed shells as coming from unspecified sandy layers cropping out between the ORTO 1 and KYLA3a samples of Dell'Angelo et al. (2012) in the Altavilla succession (Altavilla Milicia, NW Sicily, Italy). The age of these samples, presumably Late Zanclean— early Late Piacenzian, is from Dell'Angelo et al. (2012).
The “Middle Pliocene” age indicated by the respective collectors for some sites, following the three-fold Zanclean— Piacenzian—Gelasian division, is intended to be Piacenzian, according to the recently ratified, two-fold Pliocene stratigraphy (Gibbard et al. 2010).
Most of the species discussed here are from the Late Paleogene of Europe (mostly Germany and France), where the stratigraphy of the Eocene/Oligocene boundary has been debated (see Snyder et al. 1985 for a synthetic overview). As a consequence, the attribution of the Latdorfian Stage (partially corresponding to the Late Eocene/Early Oligocene Tongrian) (Müller 2011 with references) to the Eocene or to the Oligocene Series is still controversial. In the first case, involving a two-fold Oligocene (Rupelian—Chattian), the Latdorfian, as well as at least part of the Tongrian, should be included in the Late Eocene (Priabonian). In the second case, the Lat- dorfian would be the lower part of a three-fold Oligocene, as presumably intended by several 19th century and later authors (e.g., Speyer 1864; Koenen 1867; Wiechmann 1871; Görges 1957; Anderson 1960), who discussed some of the species here studied. This matter is made harder by the gap in many Late Paleogene successions due to the marked regression recorded at the Eocene/Oligocene boundary (Cavalier et al. 1981). The International Commission on Stratigraphy (see the International Stratigraphic Chart, available at http://www.stratigraphy.org/index.php/ics-chart-timescale) has ratified a two-fold Oligocene (Rupelian—Chattian) subdivision. This stratigraphical issue is beyond the scope of the present article, for which we have considered material indicated as from Latdorf (or Latdorfian/Lattorfian) as older than that reported from Rupelian or Chattian. Material referred to middle Oligocene, following the three-fold division, therefore is Rupelian in age.
As for most of the material from the Paris Basin, the collecting sites (Jeure and other localities cited by Deshayes 1864) are from the “Sables supérieurs”, therefore from the Late Tongrian, actually (Early) Rupelian (Harris and Burrows 1891; Keen 1972). The material collected at (Château) Brunehaut (locality not cited by Deshayes 1864) is from the same age/formation (Harris and Burrows 1891). The Late Tongrian material from Klein-Spauwen “Tongrien ass. sup. Tg2b” (type locality of Rissoa duboisii Nyst, 1845) is also of Early Rupelian age (see also Anderson 1960: 25).
Class Gastropoda Cuvier, 1797
Subclass Caenogastropoda Cox, 1959
Order Littorinimorpha Golikov and Starobogatov, 1975
Family Rissoidae Gray, 1847
Subfamily Rissoinae Gray, 1847
Genus Galeodinopsis Sacco, 1895
Galeodinopsis biangulata (Deshayes, 1864)
Figs. 9, 10.
1863 Rissoa Duboisii Nyst; Sandberger 1863: 131–132, pl. 10: 10–10a, as Rissoa succincta Nyst in pl. 10 (lapsus calami).
1864 Rissoa biangulata nov. sp.; Deshayes 1864: 407–408, pl. 24: 29–30 (figures in separate volume, 1866).
1867 Rissoa Duboisii Nyst; Koenen 1867: 113.
1921 Manzonia (Taramellia) areolifera Sandberger; Cossmann 1921: 31, pl. 2: 11, 12.
1960 Alvania (Taramellia) duboisi (Nyst, 1843); Anderson 1960: 25, pl. 3: 3.
1985 Manzonia (Manzonia) duboisi (Nyst, 1843); Ponder 1985: 48, fig. 100C.
2008 Galeodinopsis duboisi (Nyst, 1843); Garilli 2008: figs. 100–103.
2011 Taramellia duboisi (Nyst, 1845); Müller 2011: 24–25, pl. 5: 22a, b.
Type material: One possible syntype in UCBL provided with original label by Deshayes (Fig. 10A, B), with catalogue number EM 31000.
Type locality: Deshayes (1864: 407) reported Jeure, Montmorency, Versailles (Paris Basin, France) and Weinheim (Mainz Basin, Germany). This last locality is with reference to the material published in Sandberger (1863: 131, pl. 10: 10–10a).
Type horizon: Deshayes (1864: 407) reported “Sables supérieurs” (upper sands), from the Paris Basin, which are Early Rupelian in age.
Material studied.—10 shells in MNHU, MB.Ga.7744, as Rissoa duboisi Nyst, Waldböckelheim, Rheinland-Pfalz, W Germany, middle Oligocene marine sand; 5 shells, MZPD Mal 2036, originally in MF, FE15A, in two separate lots as Alvania duboisi and Manzonia duboisi, Hessen, SE of Kassel, Hessisch-Lichtenau, Glimmerode, central Germany, Late Oligocene (Chattian); 28 shells in UCBL, Deshayes coll. (one of them with original label), Tertiary of the Paris Basin, France; 10 shells in UCBL, Paris Basin coll., Brunehaut, France; 18 shells, in UCBL, Paris Basin coll., Jeures, France.
Description.—Shell small, conical-ovate to conical-turrited, reaching 4.95 mm in height and 3.15 mm in width; Ht/W quite variable, also depending on occurrence and position of varices on last whorl, from 1.4 to 2 in shells over 3 mm in height, average value 1.62. Protoconch multispiral, conical with 2.2 to 2.3 rounded whorls and a rather immersed nucleus. Protoconch 1 of about 0.7 whorls with netted sculpture consisting of 6–7 very thin spiral lirae and numerous, irregular, short, very narrow axial segments occurring in interspaces between lirae. Protoconch 2 sculptured with microscopic dots roughly forming one adapical and two abapical discontinuous fine spiral threads, particularly next to protoconch/teleoconch boundary. Protoconch/teleoconch boundary well-marked by thin, sinuous lip. Teleoconch with up to 4.5 convex whorls. Whorls sculptured with axial and spiral elements forming more-or-less strong clathrate, almost bicarinate pattern and acute knobs at their intersections. Axial ribs slightly curved, opisthocline to almost straight, most more prominent than spirals; 13–16 (rarely 17–18) on the penultimate whorl; lacking on shell base. Main spiral sculpture of cords overriding axial ribs; 2 on early and penultimate whorls; 6–8 and 3–6 on last whorl and on shell base, respectively. A very weak to obsolete subsutural cord occurs next to suture in some specimens. Teleoconch microsculpture on spiral cords of fine, rather flat, closely spaced spiral threads, joined by very low, short axial elements in a few specimens. Last whorl well expanded, making up 0.6 to 0.75 of total shell height, some shells bearing one varix at angles of 65° to 340°. Aperture wide, ovate, very rounded anteriorly, with acute angle posteriorly, making up about 0.4–0.55 and about 0.6–0.75 of total shell and of last whorl height, respectively. Outer lip sinuous, weakly opisthocline, double-rimmed; internally smooth, with thin rim on its edge; externally with very marked varicose swelling. Inner lip curved, weakly expanded in columellar part, forming very narrow to obsolete umbilical chink.
Remarks.—Often misidentified as Rissoa duboisii Nyst, 1845, above all because of the low-quality original illustration and the inadequate description by Nyst (1845: pl. 37: 19), Galeodinopsis biangulata is characterized by having 2 spiral cords on all teleoconch whorls with the exception of the last. It shows marked variability in Ht/W ratio and in the robustness of spiral and axial elements. The morph with delicate sculpture (Figs. 9F and 10E) matches that illustrated by Cossmann (1921: pl. 2: 11, 12) as “Manzonia (Taramellia) areolifera (Sandb.)”, a determination probably due to a lapsus calami in Sandberger (1863). Actually, Rissoa areolifera Sandberger, 1863 is a quite different species attributed to Alvania, as shown by material from the type locality (catalogue number MB.Ga.7749 at MNHU) matching the original illustration in Sandberger (1863: pl. 10: 11–11a as “Rissoa Duboisi”, lapsus calami). In fact, this species is represented by conical shells lacking varices, and with a smaller aperture without a double-rimmed outer lip (see Fig. 7A, B).
The type specimen illustrated by Deshayes (1866: pl. 24: 29, 30) was not found in his coll. from the Paris Basin at UCBL. However, indisputable conspecific specimens in the same institution are in this and other colls. from the same geological formation. Among them one specimen (EM 31000) with an original label from Deshayes (Fig. 10A, B) could be regarded as a syntype. It has all of the main characterizing characters as illustrated by Deshayes (1866: pl. 24: 30). Another specimen from the same collection (Fig. 10E) matches well that illustrated by the same author (Deshayes 1866: pl. 24: 29). Conspecific shells (Fig. 10C) in UCBL, from “Jeures”, near Fontenay (Paris Basin coll.), could be regarded as topotype material as this locality is the same as cited in the original description (“Jeure”, see Deshayes 1864: 407). This locality is also known as “Jeurre” (Harris and Burrows 1891).
The microsculpture of Galeodinopsis biangulata is different from that shown by the type species (see below), having a sort of pseudopitted pattern. This consists of fine, rather flat, closely spaced threads covering the main spiral cords (Fig. 9B2). The pseudopitted character is due to the very obsolete or absent narrow axial bridge that form the typical pitted surface in Manzonia. Other very narrow, raised spiral threads are present in the interspaces between the main spiral cords and on the subsutural one (Fig. 9A2). Comparable microsculpture occurs in G. semperi (Wiechmann, 1871) (see the section dedicated to this species) and in rissoid species linked to Alvinia, such as Turbo zetlandicus (Fig. 5A), with which G. biangulata shares the turrited-shouldered shell shape. However, a relevant combination of characters justifies the placement of this species within Galeodinopsis: the ornamentation of the protoconch, particularly the netted protoconch 1 (Fig. 9C2), the common occurrence of varices on the last whorl, the large aperture with a double-rimmed outer lip, and the quite inflated last whorl. Therefore, the placement in Taramellia Seguenza, 1903 (Müller 2011 as T. duboisi) also should not be accepted, as this genus is a synonym of Alvinia (Ponder 1985).
Stratigraphic and geographic range.—Late Eocene/very Early Oligocene at Latdorf (Latdorfian Stage) and Magderburg, central Germany (Koenen 1867, as Rissoa Duboisii; Müller 2011, as Taramellia duboisi); Rupelian of northern France (Paris Basin) and western Germany (Mainzer Basin), at Waldböckelheim; Chattian of central, at Glimmerode, and north-central Germany, at Söllingen (Koenen 1867).
Galeodinopsis germanica sp. nov.
Etymology: From Latin germanicus, German.
Type material: The holotype (SMF 335576), represented by a presumably fully grown shell preserving protoconch and provided with two varices on the last whorl, and 6 paratypes (SMF 336787, 336788, 336789, 336790, 336791a, 336791b), originally labelled as “Alvania semperi Wiechmann”.
Type locality: Abandoned mine-pit Höllkopf, Glimmerode, south of Hessisch-Lichtenau, Hessen, Germany.
Type horizon: Early Chattian (Late Oligocene) Kasseler Meeressand.
Material.—Type material only.
Diagnosis.—Shell small, conical-ovate. Protoconch multispiral with up to 2.5 rounded whorls. Protoconch 1 with netted sculpture; protoconch 2 sculptured with fine lirae and axial segments between lirae. Teleoconch with up to 3.3 convex whorls sculptured with axial and spiral elements forming a clathrate pattern. Axial ribs slightly curved, prevailing over spirals in most specimens. Microsculpture characterized by Manzonia-like pattern, with numerous lines of pits on spiral cords. Last whorl well expanded, with one or two varices. Aperture wide, ovate, with double-rimmed outer lip. Inner lip curved, forming very small umbilical chink.
Description.—Shell small, conical-ovate, reaching 2.6 mm in height and 1.5 in width; average Ht/W of 1.65, about 1.7 in shells higher than 2 mm. Protoconch multispiral, conical, with 2.2 to 2.5 rounded whorls, nucleus rather immersed. Protoconch 1 of about 0.7 whorls with netted sculpture consisting of 5–6 very thin spiral lirae and numerous, irregular, short, very narrow axial segments in interspaces. Early protoconch 2 sculptured with 2 more-or-less discontinuous fine spiral threads in subcentral and abapical position, and very narrow, opisthocline plicae in subsutural position. Close to protoconch/teleoconch boundary only one irregular, strongly discontinuous thread formed by merged microscopic dots. Protoconch/teleoconch boundary well marked by thin, sinuous lip. Teleoconch with up to 3.3 evenly convex whorls sculptured with axial and spiral elements, forming clathrate pattern. Axial ribs slightly curved, more-or-less markedly opisthocline, more prominent than spirals; 17–22 on penultimate whorl; lacking on shell base. Spiral sculpture of almost flat cords overriding axial ribs; 4 cords close to protoconch/ teleoconch boundary, rapidly increasing to 5–6; 6–7 on penultimate whorl of shells with height over 2 mm; 11–14 on last whorl, of which 5–7 on base. Subsutural and suprasutural cords usually weaker and/or discontinuous. Microsculpture on main spiral cords of flat ridges delimited by at least 7 spiral alignments of microscopic, subcircular pores forming pitted pattern. Pores formed by extremely short axial elements in very narrow interspaces between ridges. Narrow, raised spiral ridges cover interspaces between primary spiral sculpture, formed by discontinuous, tooth-like elements, extending perpendicularly from shell surface. Microsculpture of subsutural cord lacking pitted pattern, consisting of weakly raised spiral ridges with flat centres. Last whorl well expanded, making up 0.6 to 0.72 of total shell height; some shells with one or two varices forming angles of about 100°, 180°, 250°, and 340° to plain of peristome. Aperture wide, ovate, very rounded anteriorly, with acute angle posteriorly, making up 0.4–0.5 and about 0.6–0.75 of total shell and last whorl height, respectively. Outer lip double-rimmed, slightly sinuous, weakly opisthocline, internally smooth, externally with varicose swelling covered by incremental scars and spiral sculpture. Inner lip curved, weakly expanded in columellar area, forming very narrow to obsolete umbilical chink.
Remarks.—Galeodinopsis germanica sp. nov. is easily separable from other congeners by its slender shell shape resembling some Manzonia species, with which it shares the microsculptural pattern and the sculpture with axial elements more prominent than spirals. The most similar Manzonia species is M. foraminata (Lozouet, 1998), from the French Late Oligocene (Chattian) of the Aquitaine Basin. There are, however, significant differences allowing separation between G. germanica sp. nov. and Manzonia species: the occurrence in G. germanica sp. nov. of 1 or 2 varices on the last whorl, the absence of strongly raised cords on the shell base and the less rounded and proportionally larger aperture. These are typical Galeodinopsis characters. G. germanica sp. nov. is comparable with weakly sculptured morphs of G. semperi (hereafter described), from which it differs mainly by having different sculpture, consisting of more prominent, markedly curved ribs. G. germanica sp. nov. can also be distinguished by its more rounded teleoconch whorl profile (lacking the characteristic marked subsutural ramp of G. semperi) and by the typical pitted microsculptural pattern, as in the Galeodinopsis type species, G. tiberiana (compare Figs. 2C1 and 11D4). Also it bears more spiral cords: 6–7 vs. 3–5 and 11–14 vs. 7–10 on the penultimate and last whorl, respectively. The shells illustrated by Anderson (1960: pl. 3: 1, as Alvania multicostata) and Janssen (1978a: pl. 11: 31, as Alvania semperi) possibly are G. germanica nov. sp., having convex, rounded whorls, with no or obsolete subsutural ramp, and provided with axial elements more prominent than spirals. Due to this confusion, most of the Oligocene occurrences/material reported by these authors as “A. multicostata“/”A. semperi” possibly include both G. germanica nov. sp and G. semperi (see the next section for description of this species). Unfortunately, we were unable to study this material.
Galeodinopsis germanica sp. nov. shows a sculptural pattern of protoconch 2 very similar to that observed in other Galeodinospis species, except for the occurrence of subsutural plicae (Fig. 11C3–C5). It is remarkable that the microsculpture of the very weak subsutural cords (Fig. 11D5) is similar to the microsculptural pattern seen in G. biangulata, G. semperi and other species related to Alvinia.
Stratigraphic and geographic range.—Chattian of central Germany, at Glimmerode (type locality) and Freden (Anderson 1960, as Alvania muticostata), and of western Germany, at Moers (Janssen 1978a, as Alvania semperi).
Galeodinopsis semperi (Wiechmann, 1871)
Figs. 12, 13.
1845 Rissoa Duboisii nov. sp.; Nyst 1845: 418, n° 358, pl. 37: 19.
1864 Rissoa multicostata nov. sp.; Speyer 1864: 290, pl. 41: 3–5, non R. multicostata Adams, 1850 nec R. multicostata Garrett, 1857, which belong to Rissoina d'Orbigny, 1840.
1871 Alvania Semperi nov. sp.; Schwartz v. Mohrenstern (in litt.) Wiechmann 1871: 58–60.
1954 Alvania duboisi (Nyst); Glibert and Heinzelin 1954: 351, pl. 5: 13.
2008 Alvania (Alvania) multicostata (Speyer); Müller 2008: pl. 4: 10a, b.
Type material: Repository of Wiechmann (1871) types unknown.
Type locality: Wiechmann (1871: 58–59) based his species on shells from Söllingen (type locality of Rissoa multicostata Speyer, 1864) and also cited conspecific specimens from Crefeld (also known as Krefeld), Doberg and Sternberg. All these localities, which are Chattian in age, are in N Germany.
Type horizon: Oligocene marine sand, as for Söllingen (see Anderson 1960: 24).
Material studied.—The type specimen (Fig. 12) of Rissoa duboisii Nyst, 1845 in IRSNB, 3956, Tg2b layer of Klein-Spauwen, Belgium, Tongrian; 3 shells in MNHU, MB.Ga.7745, as Rissoa multicostata Speyer, Magdeburg?, Sachsen-Anhalt, Germany, Middle Oligocene; 1 shell in MNHU, MB.Ga.43, illustrated by Tembrock (1964: pl. 7: 6) as Alvania (Arsenia?) multicostata (Speyer, 1964) var.?, from Latdorfian bed, Latdorf, Germany; 1 shell in BGR, X1239, illustrated by Tembrock (1964: pl. 7: 7) as Alvania (Arsenia?) multicostata (Speyer, 1964) var.?, drill hole Drehna 5/1960, Lusatia/Brandenburg, Germany, Late Oligocene; 1 incomplete shell in BGR, X9909, illustrated by Gründel (1997: fig. 2) as Alvania (Arsenia) semperi Wiechmann, 1871.
Description.—Shell small, conical-turrited, reaching about 4.1 mm in height and 2.6 mm in width (about 2.55 ´ 1.6 in type specimen of Rissoa duboisii); Ht/W variable, mainly depending on occurrence and position of varices on last whorl, from 1.4 to about 1.7. Protoconch multispiral, conical, with 2.3 to 2.7 rounded whorls and rather immersed nucleus. Protoconch 1 of about 0.7 whorls with netted sculpture consisting of at least 6 very thin spiral lirae and numerous, irregular, short, very narrow axial segments occurring in interspaces between lirae. Intersection of spiral and axial forming rough hexagons or squares. Protoconch 2 sculptured with very few microscopic dots roughly forming one or two discontinuous fine spiral threads in suprasutural position. Protoconch/teleoconch boundary marked by weakly sinuous, very thin lip. Teleoconch with up to at least 3.5 convex whorls. Whorls sculptured with equal axial and spiral elements forming clathrate pattern of acute nodules at their intersections. Axial ribs slightly curved, opisthocline to almost straight; 18–24 on penultimate whorl; lacking on shell base. Main spiral sculpture of cords overriding axial ribs. Cords 3–4 on early teleoconch and on penultimate whorl; 7–10 and 3–7 (depending on shell height) on last whorl and shell base, respectively. A further, much weaker, almost flat subsutural cord occurs on some specimens. Teleoconch microsculpture on spiral cords of fine, raised, closely spaced spiral threads simulating a sort of pseudopitted pattern on a few specimens. Threads more raised on subsutural cord, where they do not show any trace of pitted microsculpture. Last whorl well expanded to strongly inflated, making up about 0.7 total shell height; some shells with one varix at angles of 90° to about 200°. Aperture wide, ovate, very rounded anteriorly, with acute angle posteriorly, making up about 0.5 and about 0.6–0.7 of total shell and of last whorl height, respectively. Outer lip weakly sinuous to almost straight, double-rimmed due to a thin rim on its edge; internally smooth; externally with very marked varicose swelling covered by spiral cords. Inner lip weakly curved, with very thin expansion in columellar area, forming very narrow to obsolete umbilical chink.
Remarks.—Nyst (1845) described this species under the name Rissoa duboisii while renaming Cyclostoma scalare Dubois, 1831, preoccupied by Rissoa scalaris Michaud, 1831. However C. scalare Dubois, 1831 is a different species attributed to Manzonia (Fig. 3A; see also Kowalke and Harzhauser 2004). Therefore, the Nyst's (1845) name should not be used, in accordance with the provisions of ICZN art. 72.7, and the available name for the species discussed here is Alvania semperi Wiechmann, 1871, being Rissoa multicostata Speyer, 1864 a primary junior homonym.
The tentative placement in Arsenia Monterosato, 1884 (Tembrock 1964, as A. multicostata) is incorrect. As we have shown (Fig. 7C–E), Turbo punctura Montagu, 1803, the type species of Arsenia, is quite distant, having a different shell shape, a single-rimmed aperture and a protoconch 1 with no netted sculpture; even its microsculpture is totally different, being almost smooth. T. punctura is regarded as falling within Alvania sensu stricto, following Ponder (1985). Therefore, the attribution to Alvania Risso, 1826, as indicated by Müller (2008, 2011), also is rejected. The microsculpture of Galeodinopsis semperi is comparable to that of rissoid species such as Turbo zetlandicus Montagu, 1815 (Fig. 5A2), type of Flemellia, and is very similar to that observed in G. biangulata (compare Figs. 12I, 13A4, A5 with 9B2), although only a few specimens of G. semperi show a trace of the narrow axial bridges that form the typical pitted surface in Manzonia or in the type species of Galeodinopsis. Galeodinopsis semperi can be separated from G. biangulata by the occurrence of at least 3 cords (vs. 2) starting from the first teleoconch whorl. Another main difference is in the number of axial ribs: 18–24 in G. semperi, 13–16 in most specimens of G. biangulata. Also, the spiral cords on the penultimate and last whorls of G. semperi are more numerous. As in G. biangulata, G. semperi is characterized by a turrited-shouldered teleoconch similar to that of Alvinia. The quite inflated last whorl, bearing varices on many specimens, is the main character allowing this species to be placed within Galeodinopsis.
Galeodinopsis semperi is quite variable, above all in teleoconch sculpture, which is strongly to gently clathrate. Morphs with marked sculpture (as in the type specimen of Rissoa duboisii in Nyst coll., Fig. 12A–D, H, and in a Latdorfian shell, Fig. 13C) bear fewer cords (18 on the penultimate whorl), whereas other morphs with more delicate sculpture (Fig. 13B, D), matching the original illustration of Rissoa multicostata Speyer (1864: pl. 41: 3–5), have more spiral cords (up to 24 in the material studied) and axial ribs (4 vs. 3 in the type specimen of R. duboisii). All these morphs show the same pattern of teleoconch microsculpture (compare Figs. 12I and 13B4) and are regarded as intraspecific variability of one species. Wiechmann (1871, ex litteram from Schwartz 1864) described Alvania semperi as a slender form comparable to “Alvania multicostata Speyer” with which it shares the same reticulate sculptural pattern. This form is characterized by finer sculpture, with more numerous ribs, and by a proportionally less inflated last whorl. Koenen (1867) reported a similar form at Söllingen together with typical shells of R. multicostata Speyer, 1864. This form, possibly the same as illustrated by Müller (2011: pl. 5: 24a, b), is regarded as the expression of the wide intraspecific variability of G. semperi. Koenen (1867) and Wiechmann (1871) also noted the sculptural variability of Speyer's (1864) species, when describing a morph comparable with Rissoa partschi Hörnes, 1856, a species of Alvinia having a simple peristome (see Kowalke and Harzhauser 2004: fig. 8b). This morph, characterized by more prominent sculpture, is represented by the material described by Nyst (1845) as R. Duboisii.
The name “Rissoa Semperi Schwartz v. Mohrenstern” used by Görges (1957) with reference to Schwartz (1864, “Rissoiden, 2”) is not valid, as also stated by Anderson (1960). In fact, Schwartz (1864) never described or quoted this species/name in his Die Familien der Rissoiden.
The specimen illustrated by Janssen (1978a: pl. 11: 30) as Alvania (Arsenia) semperi Wiechmann is very close to Alvania rupeliensis Tembrock, 1964 (pl. 7: 8). However we were not able to get shells of this last species, particularly the holotype that should be housed at BGR, and indicate whether it is a well-established taxon or simply a synonymous of Galeodinopsis semperi. The worn shell illustrated by Janssen (1978a, pl. 11: 32) might belong to G. semperi.
Stratigraphic and geographic range.—Late Eocene/very Early Oligocene at Latdorf (Latdorfian Stage) and Magderburg (Tembrock 1964; Müeller 2011, as Alvania multicostata), central Germany, and Klein-Spauwen, Belgium (Tongrian Stage); Early Oligocene (Rupelian) of Belgium (Late Tongrian Stage), at Klein-Spauwen, and north-central Germany, at Magderburg; Late Oligocene (Chattian) at Doberg and Crefeld, north-western Germany (Wiechmann 1871), and Söllingen (Speyer 1864; Wiechmann 1871), north-central Germany.
Galeodinopsis tiberiana (Coppi, 1876)
Figs. 2, 14A–C, F, G.
1862 Rissoa tuba nov. sp.; Doderlein 1862: 17 (nomen nudum).
1876 Rissoa Tiberiana nov. sp.; Coppi 1876: 201–202.
1990 Manzonia fariai nov. sp.; Rolán and Fernandes 1990: 64–65, pl. 1: 4–6.
Type material: More than 100 possible syntypes (not seen) in Coppi coll., at MPOB, IPUM 13721 from the Pliocene of Emilia Romagna, N Italy.
Type locality: “Tagliata” (see Garilli 2008: 43), Gagliardella, Maranello, Modena, Emilia Romagna, N Italy.
Type horizon: Coppi (1876: 202) reported “tabiano” at “Tagliata”, which may be referred to the Zanclean—Piacenzian terrigenous succession of the Emilia Romagna Apennines.
Material studied.—Those shells listed by Garilli (2008), including 31 Recent shells in MNHN, from the coasts of West Africa, originally classified as Alvania fariai (Rolán and Fernandes, 1990); 205 shells in MF, some at MZPD Mal 2034–2035, from different Italian Neogene sites including type locality (“La Tagliata”, Maranello, Modena, N Italy), most of them originally identified as Alvania tiberiana (Coppi, 1876); 44 shells at MGUP, Doderlein coll., from the Mio-Pliocene of N Italy, originally identified as Rissoa tuba Doderlein (nomen nudum); 33 shells ex AR, as Alvania tiberiana, MZPD Mal 2032, from the Pliocene of Altavilla Milicia, Palermo, Sicily. For measurements of the shells housed in MNHN see Gofas (1999), who cited this species as A. fariai (Rolán and Fernandes, 1990).
Description.—Shell conical, sturdy, reaching 3.75 mm in height and 2.75 mm in width. Ht/W ratio 1.3–1.5. Protoconch multispiral, conical, with 2–2.2 convex whorls, nucleus rather immersed. Protoconch 1 of about 0.7–0.8 whorls, with netted sculpture, consisting of 7–8 very thin spiral lirae and numerous, irregular, short, very narrow axial segments occurring in interspaces between lirae. Protoconch 2 sculptured with very small, sparse granules, fused into 2–4 discontinuous ridges on central and abapical portions of latter part of last whorl. Some groups of granules form very short, prosocline segments on central area, mainly close to protoconch/ teleoconch transition, which is marked by thin, quite sinuous lip. Teleoconch of 3–3.5 convex whorls, sculptured with a primary pattern of strong, slightly sinuous, opisthocline ribs; 12–14 (rarely 15) on penultimate whorl, becoming very weak or lacking towards base. Ribs covered with secondary spiral sculpture, formed by narrow, flat cords; 3, 4–5 and 8–10 (rarely 11–13) cords on first, penultimate and last whorls, respectively. Much weaker subsutural cord present on some specimens. Cords bear pitted microsculpture, consisting of microscopic subcircular pores forming 8–10 spiral alignments. Numerous, closely spaced, very fine spiral ridges occur between cords, formed by rough prismatic, tooth-like elements, extending perpendicularly from shell surface. At their base, each ridge has a lamella-like expansion, which covers the interspaces between two ridges. Weak subsutural cord lacking pitted surface, consisting of raised, very fine threads. Last whorl inflated, well-expanded, 0.6 to 0.75 of total shell height, many specimens bearing 1 or 2 varices with angles from about 100° to 270°, just before the outer lip on most specimens. Aperture ovate, very rounded anteriorly, weakly angulated posteriorly, comprising about 0.40–0.55 and about 0.7–0.75 (rarely about 0.6) of total shell and last whorl heights, respectively. Outer lip sinuous, weakly opisthocline; internally smooth, with thin rim on its edge; externally with very marked varicose swelling, covered with spiral sculpture, with narrow ridge on its base, towards aperture, forming double rim. Inner lip weakly curved, with very narrow columellar thickening, forming obsolete umbilical chink. Recent specimens white. After Garilli (2008), modified.
Stratigraphic and geographic range.—Galeodinopsis tiberiana lives in the eastern Atlantic, from Mauritania to northern Angola (see also Garilli 2008, with references). In the Mediterranean area it occurs in the Miocene (Tortonian) of northern Italy (northern Apennines) and the Pliocene of northern, central and southern Italy, of southern Spain, and of Algeria. In the Atlantic, it is recorded from the Portuguese Pliocene (Piacenzian) of the Mondego Basin (Garilli 2008, with references).
Galeodinopsis cf. tiberiana (Coppi, 1876)
Figs. 14D, E, 15.
Material studied.—4 shells in MZPD (Mal 2033), ex AR, as Alvania sp., from the Pliocene yellow fine sand cropping out in the right side of the Milicia River, Altavilla Milicia, Palermo, NW Sicily, Italy.
Description.—Shell conical, more elongate than typical Galeodinopsis tiberiana, reaching 3.2 mm in height and 2.05 mm in width. Ht/W ratio of about 1.5–1.6. Protoconch as for G. tiberiana. Teleoconch of little more than 3 convex whorls, with more-or-less marked subsutural ramp. Whorls sculptured by delicate to raised ribs, becoming obsolete to lacking on last whorl, crossed by finer spiral cords. Ribs weakly opisthocline or almost straight; 18–20 on penultimate whorl. Spiral cords quite narrow, numbering 3 on very early teleoconch, 4–6 and 9–11 on penultimate and last whorls, respectively. Each cord has pitted microsculpture, consisting of microscopic subcircular pores forming 4 spiral alignments. Pitted pattern formed by few flat threads connected by very short microscopic axial elements. Microsculpture between cords as for G. tiberiana. Last whorl inflated, well-expanded, comprising about 0.7 total shell height; two shells bear a single varix with angle of 180°. Aperture ovate, very rounded anteriorly, weakly angulated posteriorly, comprising 0.45–0.5 and 0.7 of total shell and last whorl heights, respectively. Outer lip very weakly sinuous, opisthocline to almost straight; internally smooth, with thin rim on its edge; externally with marked varicose swelling, covered with spiral sculpture, with very narrow ridge on its base, towards aperture, forming weak double rim. Inner lip weakly curved, with very narrow columellar thickening, forming obsolete umbilical chink.
Remarks.—It is difficult to establish whether the material described here as Galeodinopsis cf. tiberiana really represents an extreme case of intraspecific variability. No similar morphs have been observed within the abundant material (more than 200 shells) of G. tiberiana described above, and Coppi (1876) did not mention any intraspecific variation in the original description. Agatino Reitano (Tremestieri, Catania, Italy) provided this material as collected from unspecified layers from the Pliocene succession of Altavilla Milicia, in a lot containing also several shells indisputably belonging to G. tiberiana (Fig. 14A). Among these shells, three show a characteristic galeodiniform shape (Fig. 15A1–A3, B1–B3) strongly resembling species of Alvania, such as A. carinata (Da Costa, 1778) and A. rosariae Garilli, 2008 (see Garilli 2008: figs. 1, 2, 4–8, 62–69); one shell (Fig. 14D1–D3) with more prominent sculpture might be regarded as somewhat intermediate between the galeodiniform shells and those typical of G. tiberiana.
Galeodinopsis cf. tiberiana (as illustrated in Fig. 15B1– B3) resembles Rissoa moniziana Watson, 1873 (type of Moniziella Nordsieck, 1972), which has a somewhat atypical shell shape for Manzonia, but has the typical pitted surface of the genus (see Moolenbeek and Faber 1987c: figs. 58–60; Ponder 1985: figs. 102g and h). The two species share a strongly inflated shell and their sculptural pattern, with delicate, narrow spiral cords, a common character. However, M. moniziana has a paucispiral protoconch, indicating non-planktotrophic larval development, whereas G. cf. tiberiana has a multispiral protoconch indicating planktotrophic development. Furthermore, some specimens of G. cf. tiberiana bear varices on the last whorl, a typical Galeodinopsis character.
Taxonomy and evolutionary notes.—The European Tertiary Galeodinopsis species are characterized by a conical- ovate/conical-turrited, Alvania/Alvinia-like shell shape, with a discretely to strongly inflated last whorl and a duplicated peristome outlining a wide aperture. The common occurrence of varices on the last whorl is another important character distinguishing Galeodinopsis. Whereas varices occur in galeodiniform species attributed to Alvania (as a possible case of non-phyletic convergence, see Garilli 2008), they never occur within Alvinia and they very rarely occur in Manzonia. To our knowledge, the only known case of a varicose Manzonia specimen, illustrated in the present paper (Fig. 3H), might be interpreted as a teratological case. The combination of a Manzonia-like netted protoconch 1 and pitted microsculpture possibly represents a further diagnostic character distinguishing Galeodinopsis. However, the characteristic microsculpture is manifested only from the Chattian (Late Oligocene), in Galeodinopsis germanica sp. nov., whereas the older Galeodinopsis species, G. biangulata and G. semperi (Late Eocene—Early Oligocene), have a sort of “protopitted” microsculpture that possibly represents an ancestral character. This microsculpture is due to the near absence (in G. semperi) or strong reduction (in G. biangulata) of the numerous microscopic axial ribs, forming very short bridges between the ridges of the primary spiral cords in the typical pitted pattern. This pattern is comparable to that shown by the living type species of Alvinia (A. weinkauffi), and it particularly resembles that shown by Rissoa testae (type species of Actonia) and Turbo zetlandicus (type of Flemellia). These last two species possibly are related to Alvinia, particularly the former species, which shows the typical turrited shell shape as in A. weinkauffi. It is noteworthy that both Rissoa testae and Turbo zetlandicus show characters (the netted pattern of protoconch 1, the profile of the outer lip and the duplicated peristome) comparable to those of Manzonia species, stressing a certain affinity to this genus. A sort of “protopitted” microsculpture can be seen also in Alvinia planicincta Le Renard, 1990 (see MNHN Fossil Database of Tertiary shells), which lived in the Paris Basin during the late Early Eocene (Early Lutetian), long before the first appearance of Galeodinopsis. Other Alvinia characters, such as the conical-turrited shells usually provided with a subsutural ramp, indicate that the relationship between this group and the oldest Galeodinopsis species is close. In particular, it is worth emphasizing that some of the oldest species attributed to Alvinia from the Eocene of the Paris Basin (see Le Renard 1990) resemble the typical inflated shape of Galeodinopsis. This taxonomical setting demonstrates that Alvinia is morphologically closer to Galeodinopsis than to Manzonia. However, the relationship between this last genus and Alvinia is still remarkable.
The oldest representatives of Manzonia, M. foraminata and M. moulinsi, date from the Oligocene of central western Europe, but its radiation occurred in the European Neogene, when different species lived in the Atlantic province, the Mediterranean area and the Paratethys. From the Oligocene onwards the European Manzonia species differentiated according to evolutionary trends generally involving the strengthening of axial sculpture (in most of the here considered species), a proportional decrease in peristome size and of the width/height ratio (particularly remarkable in the type species), and the remarkable strengthening of the basal cords (evident in the morphological lineage M. foraminata— M. moulinsi) (Fig. 16). Conversely, the greatest diversification within Galeodinopsis occurred in the Late Paleogene of France and Germany, with evolutionary trends involving strengthening of axial sculpture and proportional decrease in the peristome size (G. semperi—G. germanica sp. nov. lineage), on one side, and, on another side (G. semperi—G. biangulata lineage), reinflation of the last whorl (Fig. 16). As for G. tiberiana, this species shows shell characters shared by G. biangulata and G. semperi (mainly the very inflated last whorl) and by G. germanica sp. nov. (mainly the curved axials prevailing over spirals). It is therefore difficult to indicate the origin of G. tiberiana, even hypothetically. However, its pitted teleoconch microsculpture might suggest a possible lineage with G. germanica sp. nov. Particularly, G. germanica sp. nov. has (more than any other congeners) characters so close to those of the Chattian species Manzonia foraminata that the lineage G. germanica sp. nov.—M. foraminata might be interpreted as a transitional step from Galeodinopsis to Manzonia. However, the hypothesis that Manzonia derives from Galeodinopsis through this lineage should be based on a more exhaustive data set. Actually, on the basis of the material studied here, the origin of Manzonia is to be backdated to the Early Oligocene, as Manzonia moulinsi has been recorded from the Rupelian of Germany and therefore preceded the first appearances of G. germanica sp. nov. and M. foraminata (very recently described from the Chattian of the North Sea and Aquitaine Basins, respectively). However, due to the problems of classifying these fossil rissoids and the obsolete approach dedicated to this group, often badly illustrated, especially by old authors (see also Lozouet 1998, for a discussion), it seems to be implausible that these last two species may have been recorded from the Early Oligocene sites under different names.
The shells of Galeodinopsis tiberiana (AR coll.) from the Pliocene of Altavilla (Sicily) are from a mixture of samplings collected at different layers of the Altavilla Milicia succession. These shells showed such wide variation that a somewhat doubtful determination as G. cf. tiberiana was preferred for the smaller ones bearing more numerous axial ribs, which are obsolete to lacking on the last whorl. This characteristic morph, never observed from other localities, appears so distant from the typical ones that it might be interpreted as the result of incipient speciation (chronospeciation?). Detailed, microlayer-by-microlayer sampling of the Altavilla succession, allowing understanding of chronological variations through transitional morphs, possibly would help to verify this assumption.
A few living species from the Macaronesian area, Rissoa moniziana Watson, 1873 and R. spreta Watson, 1873, show some similarities to Galeodinopsis, in particular in their quite inflated last whorl, the weak basal cords and the Manzonia- like microsculpture (Moolenbeek and Faber 1987c: figs. 58–60). The former species has been regarded as belonging to Manzonia (Moniziella) (Moolenbeek and Faber 1987c); the latter has been considered a species of Manzonia (Flemellia) (Moolenbeek and Faber 1987c). The identity of Moniziella is uncertain (see Moolenbeek and Faber 1987c). However, Ponder (1985) put it in synonymy with Alvinia. We have shown that Alvinia and Flemellia lack the pitted surface; therefore the placement of R. moniziana in this group should be questioned.
Palaeobiogeography and palaeoecology.—The oldest occurrences of Galeodinopsis are from the Late Eocene/very Early Oligocene in the North Sea Basin, as testified by the records of G. biangulata and G. semperi in the Latdorfian beds of Germany and of G. semperi in the Late Tongrian of Belgium. The German part of this area, including the satellite basins (e.g., those in the Rhine Graben), was the main hosting area for Galeodinopsis species, as three of its four species lived there; G. biangulata also lived in the Early Oligocene Paris Basin (Fig. 17). The Late Eocene—Oligocene palaeogeographic reconstructions of those areas (Vincent 1990; Rögl 1999: figs. 2, 3; Rasser et al. 2008, with references) outline the North Sea and Paris Basins (N Atlantic realm) as distinct marine provinces separated by the Thule bridge, a wide land-passageway putting Great Britain in communication with central-western continental Europe until the Late Oligocene. At this time sporadic communications between the Atlantic and North Sea Basin occurred in the Channel area, as indicated by Janssen (1978b) on the basis of some Late Oligocene marine faunal affinities between the North Sea Basin and the Aquitaine Basin. However, it is possible that a southernmost connection favoured these interchanges as supposed by Rasser et al. (2008). The distribution of G. biangulata (Fig. 17) indicates that a faunal interchange (possibly unidirectional) between the North Sea and Paris Basins started in the Early Oligocene, when the Thule land-bridge was already in-filled (at least intermittently) in the Channel area, much earlier than had been supposed. The linkage between G. germanica sp. nov. and Manzonia foraminata (from the North Sea and Aquitaine Basins, respectively) suggested here, as well as the Oligocene distribution of M. moulinsi (Aquitaine and North Sea Basins), support the existence of an alternative, southern route towards the Atlantic domain of the Aquitaine Basin, likely through the Upper Rhine Graben (Fig. 17). More palaeobiogeographical data could be useful for understanding whether this interchange was unidirectional, from the North Sea Basin southwards, as suggested by the oldest occurrence of M. moulinsi in the Early Oligocene of Germany.
From a palaeocological point of view, it is remarkable that all Galeodinopsis species studied have a multispiral protoconch, indicative of planktonic larval development and, in general, of a high dispersal capability, higher than in species with direct larval development. Notwithstanding this, Galeodinopsis seems never to have lived in the eastern part of the southern margin of the North Sea Basin, nor in the Paratethyan area, even though communication between the North Sea and Mediterranean/Paratethyan areas was established during that time, especially in the Early Oligocene, when ocean circulation supplied water from the North Sea to the Paratethys (Rögl 1999: figs. 1, 2). The distributional gap in the North Sea Basin may be at least partially explained by the paralic condition that affected the southeastern part of the North Sea Basin during Late Oligocene time (Rasser et al. 2008).
Galeodinopsis tiberiana is the only Neogene Galeodinopsis, from the Mio-Pliocene (palaeo)Mediterranean and the Pliocene of Atlantic Portugal, namely areas with a tropical- subtropical climate (Monegatti and Raffi 2001). Today this species is the sole well-established living species of Galeodinopsis, having been recorded from an area (West African coasts, from Mauritania to Angola) that is constrained by an analogous climatic setting. This suggests that the extinct Galeodinopsis species also required warm climatic conditions. Tropical to warm-temperate conditions have been inferred for the Palaeogene of Europe (Vincent 1990) and, more specifically, subtropical conditions were indicated for the Late Oligocene of the North Sea Basin (Rasser et al. 2008). Actually, a warm (possibly tropical-subtropical) climatic regime in the Paleogene of the North Sea Basin is suggested by the analogies (at least at genus level) between the highly diverse molluscan assemblages from the Mediterranean Neogene and the Late Eocene—Oligocene of Germany, southwards to the Mainz basin: several molluscan taxa characteristic for warm climate, such as Bathytoma Harris and Burrows, 1891, Isognomon Lightfoot, 1786, Ficus Röding, 1798, Gemmula Weinkauff, 1875 and others, have been recorded in those areas (Janssen 1978b, 1979; Gründel 1997; Monegatti and Raffi 2001; Müller 2011). Furthermore, warm, tropical to subtropical, marine climatic conditions were inferred for this part of central Europe with particular regard to the Latdorfian faunal complex and similar assemblages of younger (Oligocene) age (Müller 2011). It is remarkable that from the middle Eocene onwards, the Cenozoic continental history of central Europe, in accordance with global marine oxygen isotope records, was characterized by the increasing degree of seasonality, and a short-term cooling at the base of the Aquitanian, approximating the Oligocene/Miocene boundary, was recorded in the Weisselster Basin, in central Germany (Mosbrugger et al. 2005). Also from a palaeogeographical point of view, the area of central-western Europe underwent remarkable changes. During the Early Miocene, marine regressive conditions rewrote the palaeogeography of the North Sea Basin, triggering the progressive reduction of its southernmost part, which was constrained to higher (presumably relatively colder) latitudes. Regressive conditions also caused the disappearance of satellite basins such as those in the Rhine Graben (e.g., the Mainz Basin; Rögl 1999). This palaeogeographic-climatic evolution was possibly responsible for the disappearance of Galeodinopsis from the mid-high latitudes of Central Europe and for its migration to the Iberian Atlantic coasts and into the (paleo)Mediterranean during Mio-Pliocene time. At the end of the Pliocene, at around 2.6 My B.P. (approximating the Piacenzian/Gelasian boundary), the marked onset of the continental ice sheet in the Northern Hemisphere (Shackleton et al. 1984) triggered stepwise climatic deteriorations at lower latitudes, causing the Mediterranean extinction of several molluscan taxa (Monegatti and Raffi 2001). The new climatic regime probably was the main cause of the disappearance of Galeodinopsis from the Mediterranean and its further migration towards the subtropical-tropical West Africa coasts, where this genus still lives.
As for the ecological/bathymetrical requirements of Galeodinopsis, it is remarkable that for the Late Paleogene around Magderburg and Glimmerode (southern part of the North Sea Basin), where G. biangulata and G. semperi lived, the palaeofauna indicates normal marine conditions in rocky-sandy shore environments (Janssen 1978b; Mueller 2008, 2011). Also, the requirements of G. tiberiana are comparable, as indicated by its occurrence in the Pliocene sands of the Mediterranean Altavilla site, the depositional setting of which was mainly referred to infra-circalittoral shelf conditions (Dell'Angelo et al. 2012).
At least four European rissoid species are identified as belonging to Galeodinopsis: G. biangulata, G. germanica sp. nov., G. semperi, and G. tiberiana. A possible incipient speciation from G. tiberiana to the morph here tentatively determined as G. cf. tiberiana should be verified by in-depth studies based on more material well dated stratigraphically.
Species of Galeodinopsis are warm, marine, shallow-water taxa, from rocky or sandy bottoms. The genus originated during the Late Eocene—Early Oligocene in the southern part of the North Sea Basin. A combination of palaeogeographical and climatic events probably played an important role in determining the disappearance of Galeodinopsis from central Europe at the end of the Oligocene, and its Neogene shifting to lower Euro-Mediterranean latitudes. Climatic deterioration events are possibly the main cause of the disappearance of this genus from the Mediterranean and the eastern Atlantic midhigh latitudes during the Pliocene and its later migration to the Recent tropical coasts of West Africa. The palaeobiogeography deduced for Galeodinopsis indicates that connections between the North Sea and Paris Basins through the Channel area commenced during the Early Oligocene. Also, a connection between the Aquitaine Basin and the southernmost part of the North Sea Basin likely formed during the Late Oligocene, possibly through the Upper Rhine Graben, as testified by the occurrence in those areas of Manzonia moulinsi.
In the light of the in-depth taxonomic data set provided here, the combination of characters such as the teleoconch microsculpture and the netted pattern of protoconch 1 represents a very diagnostic key within Galeodinopsis and Manzonia, which should be regarded as a monophyletic group much closer to Alvinia than to Alvania. As for Alvinia, species such as Turbo zetlandicus (type of Flemellia) could be regarded as belonging therein; even Rissoa testae, type of Actonia, might be included therein, although more exhaustive argumentations are needed. In particular, the chronological/ transitional lineage Alvinia—Galeodinopsis—Manzonia is supported by shell and protoconch 1 characters and, above all, by the analyzed microsculpture: raised and variably closely spaced ridges in Alvinia; wider and closely spaced ridges in the ancestral species of Galeodinopsis, flat and almost fused ridges in other Galeodinopsis and Manzonia species. Effectively, on one side, the oldest Galeodinopsis species appear to have evolved from some early Late Eocene species that could be regarded as Alvinia; the strong relationship between Alvinia and Galeodinopsis is supported by several characters shared by G. biangulata and, especially, G. semperi, and the type species of Alvinia and Flemellia. On the other side, the similarities between Galeodinopsis and Manzonia are so obvious, particularly for the hypothesized lineage G. germanica sp. nov.—Manzonia foraminata, that the latter genus appears to have evolved during the Oligocene from the former rather than from Alvania/Alvinia. Therefore, Galeodinopsis can be seen as a taxon intermediate between Alvinia and Manzonia.
The only well-established living species of Galeodinopsis, G. tiberiana, could be seen as a very interesting tool for investigating the supposed lineage Alvinia—Galeodinopsis—Manzonia. In particular, more independent characters, such as those based on genetic and anatomical comparisons between Manzonia species (such as the Euro-Mediterranean type species M. crassa), Alvinia species (such as the NE Atlantic-Mediterranean species A. weinkauffi and A. zetlandica) and the West African species G. tiberiana could be extremely interesting for this purpose. Similar studies on the Macaronesian Rissoa moniziana and R. spreta, which show intermediate character between Manzonia and Galeodinopsis, would be useful as well to shed light on a number of living species of Galeodinopsis.
Thanks to Martin Aberhan (MNHU), Angela Ehling (BGR) and Ronald Janssen (SMF) for the loan of precious material. Martin Aberhan also provided images of our Fig. 13C1, C2 and an excellent copy of the plates in Sandberger (1863). Thanks to Agatino Reitano (Tremestieri Etneo, Italy) and Maurizio Forli (Prato, Italy) for providing very useful material; to Marco Taviani (Istituto di Scienze Marine, CNR, Bologna, Italy) for making available subsamples from the oceanographic cruise CS'96. Joachim Gründel (Institut für Geologische Wissenschaften, Fachrichtung Paläontologie, der Freien Universität Berlin, Germany), Ronald Janssen, Arnold Müller (Institut für Geophysik und Geologie Universität Leipzig, Germany) and Barbara Wolf (Institut fuer Geowissenschaften der Universität Kiel, Germany) provided useful literature. Our special thanks to Annelise Folie (IRSNB), who provided SEM images of the type specimen of Rissoa duboisii from Nyst coll. with copy of the original label (our Fig. 12). Arlette Armand (UCBL, U.F.R. Sciences de la Terre, Lyon, France) provided SEM images of material at Deshayes and Paris Basin colls. Lydie Touret (MINES ParisTech, Centre de Géosciences, Paris, France) and Kathie Way (Natural History Museum, London, United Kingdom) gave us information on the Deshayes collection. Abel Prieur (UCBL) kindly provided assistance during the VG visit to the Paris Basin and Deshayes collections at UCBL. We are grateful to José Leal (editor of The Nautilus, Sanibel, USA) for allowing the republishing of our Figs. 2, 3F, G, 9A1, B, and 14B, C, F, G. We are indebted to Gabriele Garilli (Palermo, Italy), who helped with translation of most of the German texts cited in this article. The online Biodiversity Heritage Library (at http://www.biodiversitylibrary.org/) is acknowledged for making available indispensable old and rare texts. This article has benefited from constructive comments by Alan G. Beu (GNS Science, Lower Hutt, New Zealand), Serge Gofas (University of Málaga, Spain), R. Janssen and Mathias Harzhauser (Natural Naturhistorische Museum Wien, Austria), and from the kind editorial advice and helpful comments. Antonio Bonfitto (Dipartimento di Biologia Evoluzionistica Sperimentale, University of Bologna, Italy), Enrico Bellia (Dipartimento di Biologia Animale, University of Palermo, Italy), R. Janssen, and Emmanuel Robert (UCBL) provided catalogue numbers of lots/specimens housed in MZP, SMF, MZPD, and UCBL, respectively.