Late Messinian rodents from Verduno (Piedmont, NW Italy): Biochronological, paleoecological and paleobiogeographic implications

Geobios 46 (2013) 111–125

Available online at

www.sciencedirect.com

Original article

Late Messinian rodents from Verduno (Piedmont, NW Italy): Biochronological, paleoecological and paleobiogeographic implications§ Simone Colombero a,*, Edmondo Bonelli b, Tassos Kotsakis c, Giulio Pavia a, Marco Pavia a, Giorgio Carnevale a a

Dipartimento di Scienze della Terra, Universita` degli Studi di Torino, Via Valperga Caluso, 35, 10125 Torino, Italy Museo Civico ‘‘Federico Eusebio’’, via Vittorio Emanuele, 19, 12051, Alba (CN), Italy c Dipartimento di Scienze geologiche, Universita` di Roma3, L.go S. leonardo Murialdo 1, 00146 Roma, Italy b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 1 December 2011 Accepted 11 October 2012 Available online 20 January 2013

The stratigraphic and paleoenvironmental context of the Verduno fossil vertebrate locality is discussed herein based on its rodent record. The Verduno section crops out in the southern part of the Tertiary Piedmont Basin (TPB), and can be included in the Messinian post-evaporitic Cassano Spinola Fm., chronologically corresponding to the so-called Lago-Mare event. Rodents are represented by a relatively rich assemblage. Murids are by far the most diverse and abundant, with at least four taxa, including the common Centralomys benericettii and Paraethomys meini, and the rare Apodemus gudrunae and Occitanomys sp. Cricetids are represented by a single species, Apocricetus cf. A. barrierei. Muscardinus aff. M. vireti appears to be the only glirid present at Verduno. The Verduno rodent assemblage shares some taxa with other Messinian post-evaporitic localities from Italy bearing continental vertebrate remains, such as Brisighella (central Italy) and Moncucco Torinese (NW Italy) (e.g., C. benericettii, P. meini) and, possibly, with Ciabo`t Cagna (NW Italy). However, the general structure of these four Messinian assemblages displays substantial differences, which may reflect different palaeoenvironmental conditions. ß 2013 Elsevier Masson SAS. All rights reserved.

Keywords: Rodents Taxonomy Biochronology Stratigraphy Paleoenvironment

1. Introduction The present study is part of a broader research work focused on the analysis of the structure and composition of the Late Miocene vertebrate communities of the Piedmont region. The modern stratigraphic study of the Messinian deposits of the Tertiary Piedmont Basin (TPB), as well as the analysis of the geodynamic processes that controlled the sedimentation in the basin before, during and after the Messinian Salinity Crisis, were started by Sturani in the 1970s. From the beginning of such stratigraphic researches, continental deposits were identified at the top of the Messinian succession. These deposits are characterized by the presence of brackish molluscs (the so-called Congeria beds; Sturani, 1973, 1976) and have been traditionally assigned to the Cassano Spinola Conglomerates Fm. (Boni and Casnedi, 1970; Ghibaudo et al., 1985; Clari et al., 2008; Bernardi et al., 2010; Dela Pierre et al., 2011). Messinian remains of terrestrial vertebrates are extremely rare in Piedmont, up to date known exclusively from the site of Ciabo`t Cagna, near Corneliano d’Alba (Cavallo et al., 1993).

§

Corresponding editor: Gilles Escarguel. * Corresponding author. E-mail addresses: [email protected], [email protected] (S. Colombero). 0016-6995/$ – see front matter ß 2013 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.geobios.2012.10.007

Extensive exploration of the Messinian deposits of the TPB, carried out in the last few years by the paleontologists of the Universita` degli Studi di Torino, led to the discovery and study of two new sites, Moncucco Torinese and Verduno, in the northern and southern part of the TPB, respectively. The vertebrate assemblage of Moncucco Torinese has been the subject of a preliminary analysis (Angelone et al., 2011), which evidenced the presence of a relatively large number of taxa, including several small mammals. The site of Verduno is known just for the presence of a few carnivoran remains (Hyaenictitherium sp. and Eucyon monticinensis) randomly collected along the banks of the Tanaro River (Sardella, 2008). The purpose of this paper is to present the rodent remains collected in this site during the excavation campaign organized in July 2010 by the Dipartimento di Scienze della Terra, Universita` degli Studi di Torino, and to discuss their stratigraphic and paleoenvironmental implications. Besides rodents, the other vertebrate remains collected at Verduno during the excavation campaign are currently under study and will be described in a forthcoming paper. 2. Geographic and geological settings The site of Verduno is located in the southern part of the TPB near the town of Alba, in the Cuneo Province (Fig. 1). The

112

S. Colombero et al. / Geobios 46 (2013) 111–125

Fig. 1. Geographic setting. The arrow indicates the position of the Verduno fossil site. TH: Torino Hill; MO: Monferrato; TBP s.s.: Tertiary Piedmont Basin sensu stricto. From Angelone et al. (2011).

fossiliferous deposits are restricted to the upper portion of a succession, known in the literature as the ‘‘Pollenzo section’’ (Bernardi et al., 2010; Dela Pierre et al., 2011; Fig. 2), well exposed along the Tanaro River. The Pollenzo section is part of the Alba succession and documents the sedimentary processes that occurred during the Messinian at the boundary between marginal and basinal areas in the southern portion of the TPB (Dela Pierre et al., 2011). Overall, the Pollenzo section has provided relevant data for the improvement of the knowledge of events related to the Messinian Salinity Crisis (Bertini and Martinetto, 2011; Dela Pierre et al., 2011). The base of the Pollenzo section consists of marine preevaporitic sediments of the Marne di S. Agata Fossili Fm., which are overlain by the evaporites of the Alba Primary Lower Gypsum Unit, corresponding to the Vena del Gesso Fm., formerly known as the Gessoso-Solfifera Fm. (Roveri and Manzi, 2007). This unit consists of euxinic shales and different types of gypsum lithofacies that testify the evaporitic interval of the Messinian Salinity Crisis (5.96– 5.55 Ma; CIESM, 2008; Dela Pierre et al., 2011). The basal portion of the post-evaporitic deposits corresponds to the Valle Versa Chaotic Complex, characterized by slumped mudstones with displaced meter-sized gypsum slabs, whereas the upper portion consists of fresh- and brackish-water deposits (muddy and silty beds crossed by sandy and gravelly layers) of the Cassano Spinola Conglomerates Fm. (Dela Pierre et al., 2011). This unit was deposited in the terminal phase of the Messinian Salinity Crisis, approximately from 5.5 to 5.33 Ma (CIESM, 2008; Dela Pierre et al., 2011). The uppermost layers of the Cassano Spinola Conglomerates Fm. are characterized by greenish marls whose upper part progressively becomes more bioturbated (Bernardi et al., 2010). This unit is followed by a 30–40 cm-thick muddy black layer. Many traces of bioturbation are present in this level, such as galleries filled by gray marls of the overlying Argille Azzurre Fm. Because of the presence of the nannofossil species Reticulofenestra zancleana, the lowermost portion of the Argille Azzurre Fm. of the Pollenzo section can be assigned to the lower Zanclean MNN12a calcareous nannofossils subzone (Bernardi et al., 2010).

3. Stratigraphy of the excavation site The excavation site is located approximately 20 m below the ‘‘black layer’’ marking the Mio-Pliocene boundary, and 25 m above the slumped marls of the Valle Versa Chaotic Complex, thereby implying that the vertebrate-bearing layers clearly belong to the Cassano Spinola Conglomerates Fm. It is very difficult to define in great detail the exact stratigraphic position of these fossiliferous layers within the section because of the inadequate knowledge of the stratigraphic and sedimentological features of the Cassano Spinola Conglomerates Fm. at Pollenzo which is also affected by faulting, and, more importantly, because the upper part of this section is covered by a very thick layer of deposits derived from the annual floods of the Tanaro River. Five distinct stratigraphic layers can be recognized in the excavation site, with a maximum thickness of about 2 m (Fig. 3). The three lower layers can be followed laterally for tens of meters as they crop out in the bed of the Tanaro River. The two upper layers show the typical features of the fluvial facies (Einsele, 1992) and are laterally discontinuous, following the erosive surfaces over which they lay on. The fossil remains come from layers 1, 2 and 5. The succession is schematically summarized on Fig. 3 and can be described as follows from top to bottom:  layer 5 (20–50 cm) lies on an erosive surface and consists of cross-bedded conglomerates with subcentimetric pebbles, mainly of gypsum. This layer is separated from layer 4 by an erosive surface. Fossils are represented by remains of small mammals;  layer 4 (50–90 cm) is characterized by sub-horizontal or, more rarely, oblique laminated greenish to gray sands. The base is made of by conglomerates. These barren sediments lay over an erosive surface. A portion of this layer is covered by the current deposits coming from Tanaro river flooding;  layer 3 (10–15 cm) consists of gray to greenish clays with subhorizontal or slightly oblique, partially laminated silty clays. There are no fossils;

S. Colombero et al. / Geobios 46 (2013) 111–125

113

glirids. Four species of murids are present. Two of them are extremely common and represent, by far, the largest part of the available small mammal material. The examined material is currently housed in the Museo Civico ‘‘Federico Eusebio’’, Alba, Italy. 5. Systematic paleontology

Fig. 2. The Pollenzo section. The arrow and the box indicate the stratigraphic position of the excavation area. Lith. unit: lithostratigraphic unit; Zancl.: Zanclean; AAF: Argille Azzurre Formation; VVC: Valle Versa Chaotic Complex. From Dela Pierre et al. (2011).

 layer 2 (10–15 cm) is composed by dark-brownish to black clays with small sandy conglomeratic lenses at the base. Ramified burrows filled with gray to greenish sediments are present at the top of the layer. The fossil content, consisting of small mammal remains and mollusk shells, is very abundant. This layer appears to be separated from the lower one by a hardly distinguishable erosive surface placed in correspondence with the conglomeratic lenses;  layer 1 (50 cm) consists of gray to greenish clays with subhorizontal, rarely oblique, partially laminated silty clays. About 40 cm below the base of layer 2, some remains of large mammals, currently under study, were found. Some small mammal remains are also present, including partially articulated skeletal elements. 4. Material Approximately 1 ton of sediment from the fossiliferous levels of the excavation site of Verduno was washed and sieved. Rodents are very abundant. They are represented by murids, cricetids and

Order RODENTIA Bowdich, 1821 Family MURIDAE Illiger, 1811 Genus Centralomys De Giuli, 1989 Centralomys benericettii (De Giuli, 1989) Fig. 4(D–F, M–O) Material and measurements: Table 1. Description: M1: Tubercles are slender, pointed at the top and slightly widen at the base. Cusps of the labial side (t3–t6–t9) are well aligned, forming a roughly rectilinear labial margin, as well as the two cusps of the lingual margin (t1–t4). The t1 is located well behind t2 and very close to t5; it is inclined posteriorly and internally towards the t5. It is always connected to t5; such a connection can be attained through a short longitudinal spur extending from the posterolabial margin of the cusp or, alternatively, through the posterior margin of the cusp due to close proximity of t1 and t5. A t1bis is always present; it generally develops from the lingual side of t2 and then connects to the t1. The t2 is large and stout. A t2bis is present only in rare cases. The t3 is located in a more advanced position with respect to t1; it exhibits longitudinal spurs that projected posteriorly. Approximately 70% of the specimens are characterized by spurs that barely touch the base of t5; cusps t4–t5–t6–t9–t8 form a complete stephanodont crown; t4–t8 connection, however, is generally low. The t4 is located in a backward position. In some worn specimens the t4 is directly in contact with the t8. The t5 is large and stout. The t6 is strongly inclined backwards; it is the most developed cusp of the labial side; it is connected to t5 and t9. The t8 is usually the largest cusp of the tooth. The t9 is smaller than t6. The t12 is present and usually occurs as an angular extension that develops from the t8. M2: Cusps are slender, pointed at the top and slightly widened at the base. The t1 is vertically developed; it presents short extensions towards the t5. A t1bis is always present, generally smaller than t1; it is placed slightly more labially than the t1; connected to the anterior margin of the t1 with which gradually melts at advanced stages of wear; very close to the anterolingual margin of t5 that eventually contacts. The t3 is smaller than t1, similar in size to t1bis; it is located near the anterior margin of the molar. Tubercles t4–t5–t6–t9–t8 form a stephanodont crest; the connections between t4–t6 and t8–t9, however, are low, reaching about half of the development of the tooth crown. The t4 is as large as or larger than t1; it is placed at the same level as the t6–t9 connection, and connected to t9 only through a low ridge. The t5 is large and stout. The t6 is large; it is in a more advanced position than t4. The external margin of this tubercle is rather sharp. The t9 is smaller than t6 to which it is particularly close. The connection with the t8 is attained through a ridge that reaches approximately the half of the height of the tooth crown. The t8 is large and stout. The t12 is reduced. There is a small extension in the direction of the labial enamel, which may be absent in some worn specimens. Three or four (15%) roots may be present. M3: The t1 is large; it extends labially toward the anterior margin of t5. The contact can be observed at higher stages of wear. The t3 is absent. Tubercles t4–t5–t6 converge to form a chevron. The t8 seems to be isolated in unworn specimens, while, at average stages of wear, a connection with the t6 and, more rarely, the t4 can be observed. The molar has three roots. m1: Cusps become rounded slightly towards the base. A tiny tma is visible in some specimens (less than 35%). The anteroconid

114

S. Colombero et al. / Geobios 46 (2013) 111–125

Fig. 3. Excavation area of Verduno in the context of the Cassano Spinola Conglomerates, Pollenzo section. A, I: the white lines highlight the contact between layers 1, 2 and 3. A, II: the white line highlights the erosive surface between layer 4 and the top of layer 2; layer 3 is absent in this sector due to the erosion. A, III: the white line highlights the erosive surface between layers 4 and 5 in the uppermost part of the section. B: stratigraphical column of the excavation area.

complex is slightly asymmetric, with the lingual cusp in a more advanced position than the labial. The first two series of cusps (anteroconid complex and protoconid-metaconid) are always in touch. In the specimens in whom the wear condition permits, we can observe that the contact is more developed on the lingual side even if a specimen presents a best-developed contact on the labial side and another equally-developed on both sides. The metaconid and protoconid are slightly inclined forwards and the complex is slightly asymmetrical with the metaconid being placed in a more advanced position. The hypoconid and the entoconid form an asymmetric cordiform complex with the entoconid in a more advanced position than the hypoconid. The cusps are slightly inclined forward. A longitudinal spur extends usually towards the postero-lingual edge of the protoconid, very close to the junction with the metaconid. In some cases (50%) the spur develops in a narrow ridge. The posterior heel is present but not well developed, in some cases absent or only sketched. It is oval-shaped, slightly shifted on the lingual side. In advanced stages of wear it can be noted that it usually connects to the hypoconid and more rarely to the entoconid. The posterior accessory cusp (c1) is well developed

and sometimes it connects to the hypoconid. From it a labial cingulum develops, connecting the anteroconid. Generally, the cingulum is not in contact with the protoconid. Some cusplets can develop occasionally on the labial cingulum. There are two roots, and rarely a slight hint of a third central radicle. m2: Cusps are slender. The anterolabial cusp is well developed; at the base it extends to the protoconid, with which it can be rarely connected. Anterior cusps form a reniform complex and are slightly inclined anteriorly. The two posterior cusps (entoconid and hypoconid) form a cordiform complex with the entoconid in a more advanced position. From this complex a longitudinal spur extends toward the base of the anterior complex; in some cases (about 50%) the longitudinal spur develops in a narrow ridge. A low labial cingulum is present, developing from the anterolabial cusp to the posterior accessory cusp (c1). This cusp is small and connected to the hypoconid; in some cases it can be absent. Sometimes the labial cingulum is under-developed and weakly visible. In other cases it may present small cusplets formed only by slight swellings on the enamel. The posterior heel is well formed, oval and strongly compressed, slightly shifted to the lingual side. There are two roots.

S. Colombero et al. / Geobios 46 (2013) 111–125

115

Fig. 4. A–C, J–L. Paraethomys meini: A, right M1, V017; B: right M2, V040; C: left M3, V204; J, right m1, V209; K, left m2, V042; L, left m3, V201. D–F, M–O. Centralomys benericettii: D, left M1, V072; E, left M2, V099; F, right M3, V074; M, right m1, V030; N, right m2, V120; O, right m3, V126. G, H, P, Q. Apodemus gudrunae: G, left M1, V009; H, left M2, V067; P, right m1, V271; Q, left m2, V272. I. Occitanomys sp.: left M2, V290. Scale bar: 1 mm.

S. Colombero et al. / Geobios 46 (2013) 111–125

116

Table 1 Material and measurements of Centralomys benericettii from Verduno. Dental elements

Length

Width

TNr

mNr

min

mean

max

s

mNr

min

mean

max

s

1

M1 M2 M3 m1 m2 m3

11 8 1 4 4 2

10 7 1 4 4 2

1.67 1.22 – 1.52 1.14 0.93

1.76 1.27 0.84 1.60 1.21 0.94

1.86 1.32 – 1.65 1.25 0.95

0.0583 0.0365 – 0.0621 0.0463 –

9 7 1 4 4 2

1.19 1.18 – 0.97 1.00 0.86

1.23 1.22 0.91 1.01 1.10 0.87

1.28 1.27 – 1.05 1.14 0.87

0.0322 0.0330 – 0.0424 0.0639 –

2

M1 M2 M3 m1 m2 m3

24 19 4 28 22 12

19 16 4 21 20 11

1.65 1.15 0.86 1.50 1.15 0.90

1.76 1.23 0.92 1.62 1.22 0.97

1.87 1.33 0.98 1.73 1.30 1.01

0.0700 0.0676 0.0544 0.0696 0.0373 0.0315

18 16 4 21 19 11

1.16 1.11 0.83 1.00 1.05 0.83

1.23 1.20 0.92 1.07 1.13 0.88

1.32 1.32 0.97 1.18 1.26 0.95

0.0440 0.0540 0.0646 0.0416 0.0578 0.0340

5

M1 M2 M3 m1 m2

5 3 1 3 10

5 3 1 3 10

1.77 1.33 – 1.59 1.10

1.79 1.34 0.90 1.63 1.25

1.86 1.34 – 1.71 1.33

0.0371 0.0058

5 3 1 3 10

1.22 1.10 – 1.10 1.03

1.27 1.20 0.91 1.11 1.14

1.32 1.27 – 1.12 1.24

0.0371 0.0907 – 0.0100 0.0606

Layer

0.0666 0.0649

Layer: stratigraphic position of the remains (see Fig. 3); TNr: total number of identified specimens for each dental element; mNr: number of measurable specimens (length or width) for each dental element; s: standard deviation.

m3: There is a small anterolabial cusp formed by a swelling of the enamel. Protoconid and metaconid are associated in a symmetrical reniform complex that can be slightly inclined posteriorly. The posterior complex presents a very compressed and transversely elongated, semicircular shape; it is slightly inclined anteriorly and slightly shifted on the lingual side, and can be connected to the metaconid at high level of wear. Remarks: The remains of C. benericettii from Verduno were compared to specimens of C. benericettii from the Monticino quarry (RA, Central Italy). The average size of the population of Verduno falls within the size range of the type population of C. benericettii from the type locality Brisighella 1 (BRS 1; De Giuli, 1989), and Brisighella 25 (BRS 25; Martı´n-Sua´rez and Mein, 1991). The morphology of the specimens from Verduno is similar to that of the type population of BRS 1, even if some slight differences occur: remains from Verduno exhibit a slightly higher degree of stephanodonty, especially in the development of the posterior spurs of t3 as well as in the connections between the last two complexes of cuspids in the lower molars. Nonetheless, the population of C. benericettii from BRS 25 exhibits morphological features reminiscent to those of the population of Verduno, characterized by a similar degree of stephanodonty. The average size of the teeth of C. benericettii from Verduno is slightly smaller than that of Occitanomys debruijni (Sen, Jaeger, Dalfes, Mazin and Bocherens, 1989) from Maritsa (Sen et al., 1989) and Tomea Eksi 1 and 2 (Hordijk and de Bruijn, 2009). We compared the material from Verduno with some teeth of O. debruijni (Sen et al., 1989) from Maritsa (Greece, MN14) housed in the collection of the University of Lyon 1. C. benericettii differs in some morphological characteristics. In particular, the t1 of M1 of O. debruijni is often well separated from the t2, whereas the separation of these tubercula occurs very rarely in the specimens from Verduno. Moreover, the connection t4–t8 is lower in specimens of O. debruijni. In addition, the connections t3–t5 are less developed than in the material from Verduno. Lower molars usually have a less developed labial cingulum, although this character may exhibit a great variability (Martı´n-Sua´rez and Mein, 1991). Specimens of C. benericettii from Verduno are slightly smaller than those of Occitanomys alcalai Adrover, Mein and Moissenet, 1988 from different localities of Spain (Adrover et al., 1988, 1993; Garcı´a-Alix et al., 2008b; Minwer-Barakat et al., 2009a, 2009b),

even if the size ranges can partially overlap. However, they differ in some morphological characteristics. In particular, in the M1 of O. alcalai, the longitudinal spurs that extend from the t3 are less frequent and never reach the t5. Moreover, the t1 of O. alcalai seems to be placed in a slightly more advanced position. In the m1 of C. benericettii, the longitudinal spurs seem to be more developed. According to the description of the type population of Peralejos E (Adrover et al., 1988), the m1 never exhibits the tma even if this character can rarely occur in specimens from other locations such as Valdecebro 6 and La Gloria 5 (Adrover et al., 1993), and Negratı´n-1 (Minwer-Barakat et al., 2009a). The remains of C. benericettii are larger than those of Castillomys gracilis van de Weerd, 1976 from many localities, including Caravaca, Orrios (van de Weerd, 1976), La Gloria 4 (Adrover et al., 1993), Purcal 13, Calicasas 3 and Calicasas 4B (Garcı´a-Alix et al., 2008b). They also differ from the less-developed stephanodont crown that characterizes the teeth of C. gracilis, in which the t4 is isolated from the t8. In the M2 of C. Gracilis, the t1bis is absent whereas it is always present in C. benericettii. Moreover, the labial cingulum of the m1 is less developed in C. gracilis than in C. benericettii. Genus Paraethomys Petter, 1968 Paraethomys meini (Michaux, 1969) Fig. 4(A–C, J–L). Material and measurements: Table 2. Description: M1: Cusps are large and bulky. The enamel is generally well developed, providing strength to the molars. The t1 is large; it is placed in a particularly backward position. Generally, the connection t1–t5 is not present, but in much worn specimens it can be observed that a short spur extends towards the t4–t5 connection. The connection with the t2 is weak but always present and observable even at a low level of wear. The t1bis is absent. The t2 is large. The t3 is smaller than t2 and t1; it is very close to t2 and placed in a more advanced position with respect to t1. The connection with t2 is observable at a low stage of wear. A posterior spur extends from t3 towards the t5–t6 connection, never reaching it. The t4–t5–t6–t9–t8 are united to form an almost continuous stephanodont crown in all specimens. The t4 is voluminous, with the same size of t1, set in a rather rearward position, approximately at the level of the t6–t9 connection. The base is situated very close to t8, but in unworn specimens it can be noted that the

S. Colombero et al. / Geobios 46 (2013) 111–125

117

Table 2 Material and measurements of Paraethomys meini from Verduno. Dental elements

Length

Width

TNr

mNr

min

mean

max

s

mNr

min

mean

max

s

1

M1 M2 M3 m1 m2 m3

6 2 1 5 2 1

6 2 1 5 2 1

2.33 1.59 – 1.97 1.61 –

2.38 1.64 1.05 2.11 1.62 1.38

2.46 1.68 – 2.17 1.63 –

0.0552 – – 0.0827 – –

6 2 1 5 2 1

1.60 1.50 – 1.24 1.47 –

1.63 1.56 1.11 1.36 1.48 1.17

1.68 1.62 – 1.44 1.48 –

0.0316 – – 0.0772 – –

2

M1 M2 M3 m1 m2 m3

20 7 3 8 12 6

10 6 3 5 9 4

2.19 1.56 1.14 2.05 1.31 1.20

2.32 1.68 1.19 2.14 1.59 1.28

2.49 1.83 1.24 2.20 1.69 1.36

0.1001 0.9800 0.0503 0.0560 0.1112 0.0679

14 4 3 7 9 6

1.49 1.47 1.18 1.34 1.40 1.16

1.61 1.60 1.22 1.39 1.48 1.22

1.69 1.70 1.29 1.47 1.57 1.29

0.0633 0.0946 0.0624 0.0472 0.0580 0.0468

5

M1 M2 M3 m1 m2 m3

3 3 1 1 4 1

3 3 1 1 3 1

2.39 1.61 – – 1.54 –

2.45 1.65 1.15 2.21 1.58 1.30

2.56 1.69 – – 1.64 –

0.0954 0.0404 – – 0.0513 –

3 3 1 1 4 1

1.64 1.60 – – 1.46 –

1.65 1.62 1.18 1.44 1.50 1.14

1.67 1.64 – – 1.53 –

0.0153 0.0208 – – 0.0287

Layer

Layer: stratigraphic position of the remains (see Fig. 3); TNr: total number of identified specimens for each dental element; mNr: number of measurable specimens (length or width) for each dental element; s: standard deviation.

contact between these two cusps is very low and only occurs through the enamel, or with a very low and thin ridge. The t5 is large; it has well-developed connections with both t4 and t6. The t6 is the most developed cusp of the labial side from which it can protrude significantly. The t9 is reduced, with the same size of t3 or even smaller. It is connected with t6 and t8. The t8 is large. The t12 is poorly developed, represented by an extension of the enamel of the t8. M2: The t1 is well developed. A short low spur extends posteriorly towards the t5, rarely reaching the base of this cusp. The t3 is much smaller than t1; it does not contact t5. The t4–t5– t6–(t9)–t8 are united to form an incomplete stephanodont crown, the connection between t4 and t8 being very low or absent. The t4 is slightly larger than t1; it presents a good connection with the t5. The connection with the t8 is very low or absent in some cases; it is placed at about the same level than t6. The t5 is large; it is characterized by wide connections with t4 and t6. The t6 is slightly smaller than t4; it is located slightly posterior to the t5. The t9 is absent or poorly developed; when absent, a long and narrow ridge connects directly the t6 and t8. In rare cases, however, a small cusp is developed; in other cases, the ridge bears slight bulge. The t8 is the most voluminous tubercle; it is strongly inclined posteriorly. The t12 is absent. Three roots are present. M3: The shape of the molar is approximately triangular. The t1 is well developed; it may exhibit very short extensions towards the anterior border of the t5, from which it is always isolated. In a few specimens there is a slight bulge on the enamel in the position of the t3. The t4–t5-t6 form a chevron. The lingual side of the chevron (t4) is slightly longer than the labial one (t6). The complex formed by t8–t9 has a semicircular section. In unworn specimens there are not visible contacts with the anterior chevron. Nonetheless, in some slightly worn specimens it can be noted that a contact occurs through the enamel, in particular with the t4 on the lingual side. More rarely, an even more weak contact can develop with the t6 so as to surround a circular depression. There are three roots. m1: The molar has a robust structure. The cusps are large and stout. In very few specimens a small tma is present, formed by a slight swelling of the enamel. The anteroconid complex is asymmetric, with the lingual lobe placed in a more advanced position. The cusps of the lingual side are similar in size to those of the labial side. The second set of cusps (metaconid and protoconid)

is totally separate from the anteroconid complex in around 20% of the specimens. Another 20% of the specimens show a contact primarily with the metaconid on the lingual side, while in other specimens (ca. 20%) it occurs mainly with the protoconid on the labial side. In the remaining specimens (ca. 40%), the contact occurs both on the labial and lingual sides; cusps are slightly inclined towards the anterior border and are united to form a reniform complex. Hypoconid and entoconid are associated to form a reniform or cordiform complex. From this complex a short longitudinal spur may extend towards the protoconid-metaconid complex. However, the cusps remain always isolated although in much worn teeth it can be noted that the spur almost reaches the base of the metaconid. The posterior accessory cusp (c1) is always present and well developed. It is located in a slightly more advanced position than the hypoconid with which it is generally connected. A second accessory cusp is generally present, developing near the protoconid. Sometimes, this cusp bears a very low anterior extension that can reach the labial lobe of the anteroconid complex. A third and a fourth cusplet may also develop on the labial cingulum. The posterior heel is always present, generally elliptical in shape, slightly shifted towards the lingual side; usually, it exhibits a small size. There are two roots. m2: Cusps are robust. The anterolabial cusp is large; it develops in a particularly advanced position always ahead of the anterior margin of the protoconid. The contact with the protoconid is absent or hardly observable in worn teeth. Metaconid and protoconid merge together to form a reniform complex. Cusps are inclined slightly forwards. Entoconid and hypoconid are joined to form a cordiform complex. From this complex a longitudinal spur extends towards the anterior cusps without touching them. The labial cingulum is reduced or formed by tiny cusplets. The posterior accessory cusp is usually well developed. It is situated near the hypoconid with which it can come into contact. Occasionally, another cusp of smaller dimensions is present, just below the anterolabial cusp. The posterior heel is usually elliptical in shape. There are two roots. m3: Cusps are voluminous and characterized by a thick enamel layer. The anterolabial cusp is absent. Protoconid and metaconid are joined to form a reniform complex. The posterior complex has a semicircular section; it is slightly shifted to the lingual side. A posterior accessory cusplet (c1) may very rarely develop on the

S. Colombero et al. / Geobios 46 (2013) 111–125

118

labial side, always in contact with the posterior complex. When present, it seems to be a labial extension of the hypoconid. The posterior complex might be occasionally connected to the protoconid-metaconid complex mainly on the lingual side. There are two roots. Remarks: The material of P. meini from Verduno has been compared with that from the type locality of Se`te (France). The specimens from Verduno are consistent in size and morphology with those from the type locality. However, some slight differences occur: in the M1 from Verduno, the t1 seems to be closer to the t5 and the t3 presents a more developed spur. Nonetheless, in other populations of P. meini, as for example those from some localities of the Granada basin, especially La Dehesa 4A and 4B, the spurs of the t3 are more developed (Garcı´a-Alix et al., 2008b). We have compared our material with that of other populations of P. meini (Esbarrondadoiro, Celadas 9, Serrat d’en Vaquer, Terrats, Villeneuve de la Raho, Mont-He´le`ne). Overall, all the individuals exhibit morphological characteristics very similar to those from Verduno. Nonetheless, it is worth noting that some features of the teeth of P. meini from these populations and from Verduno display a certain degree of variability, such as the size of the t3 (more or less developed), the development of the spurs of t3, the development of the t9, the degree of contact between t4–t8 and t6–t9 in the M1 and M2, and the number of cusplets on the labial cingula of m1 (generally two or more, with some specimen characterized by one). As far as regards the size, the dimensions of the teeth of P. meini from Verduno are similar to those from many localities of Europe, such as those of the Granada basin (Garcı´a-Alix et al., 2008b), Negratı´n-1 and Rambla de Chimeneas 3 (Minwer-Barakat et al., 2009a, 2009b), Mont-He´le`ne (Aguilar et al., 1991), and of the Teruel basin (Adrover et al., 1988, 1993). We compared the material from Verduno with the specimens of Paraethomys miocaenicus Jaeger, Michaux and Thaler, 1975 from Amama 2 and Albacete. The remains from Verduno are similar in size and morphology. Usually, the teeth from Verduno display a higher contact between t6 and t9 in the M1. The degree of development of the t4–t8 contact is variable in a way similar to that of the population of Verduno, as well as the degree of development of the t9 on the M2. On the basis of the comparisons with some populations of Paraethomys anomalus (de Bruijn, Dawson and Mein, 1970) from Maritsa, Caravaca, El Arquillo 4, Amama 3 and Brisighella, the material from Verduno exhibits a high degree of similarity in size and morphology, particularly in the development of the spurs of the t3 in the M1, that are frequently well developed; in the position of the t1, which is close to the t5; in the number of cusplets on the labial cingulum (generally two). With respect to Paraethomys abaigari Adrover, Mein and Moissenet, 1988 from Villalba Alta Rio (Adrover et al., 1988) and Paraethomys aff. abaigari from the Granada basin (Garcı´a-Alix et al., 2008b), the material from Verduno displays a smaller size and less developed spurs on the t3 of M1. Concerning

Paraethomys jaegeri Montenat and de Bruijn, 1976 from Gorafe 2, the teeth of this species are larger than those of our specimens. Moreover, in the M1 of P. jaegeri the t3 is in a more advanced position and displays a more evident contact with the t2. Many authors (among others, Montenat and de Bruijn, 1976; van de Weerd, 1976; Freudenthal and Martı´n-Sua´rez, 1999; Minwer-Barakat, 2005, Garcı´a-Alix et al., 2008b; Minwer-Barakat et al., 2009a, 2009b) consider P. miocaenicus and P. anomalus as junior synonyms of P. meini. However, some authors do not agree with the synonymy of these three species, or continue to treat them separately (among others, Mein et al., 1990; Adrover et al., 1993; Agustı´ and Llenas, 1996; Agustı´ et al., 2006). In our opinion the differences among these species are not relevant and we consider P. anomalus and P. miocaenicus as junior synonyms of P. meini. At the end of the Miocene and at the beginning of the Pliocene, P. meini was present in Europe both on the west and east sides of the Mediterranean area. This species was reported from many late Turolian and early Ruscinian localities of Spain (van de Weerd, 1976; Adrover et al., 1988, 1993; Castillo et al., 1990; Minwer-Barakat et al., 2005, 2009a, 2009b; Agustı´ et al., 2006; Garcı´a-Alix et al., 2008a, 2008b, 2008d), France (Mein and Michaux, 1970; Aguilar et al., 1989, 1991, 1999), and Greece (de Bruijn et al., 1970). In Italy, P. anomalus, now considered a junior synonym of P. meini, was previously reported from Brisighella (De Giuli, 1989). P. meini has a particular evolutionary and palaeobiogeographic meaning as, according to some authors (Garce´s et al., 1998; van Dam et al., 2001; Agustı´ et al., 2006), it is regarded as an African immigrant together with some other taxa such as camelids (Agustı´ et al., 2006). The most ancient record of this species in Europe is found in Spain at about 6.1 Ma (Garce´s et al., 1998). Genus Apodemus Kaup, 1826 Apodemus gudrunae van de Weerd, 1976 Fig. 4(G H, P, Q) Material and measurements: Table 3. Description: M1: The t1 is smaller than or equal to t2; it is slightly inclined inwards, and positioned very close to the t5. The connection between t1 and t2 is generally observable at a medium stage of wear. The t2 is large. The t3 is about the size of t1 or slightly smaller; it is placed in a more advanced position than the t1. Short spurs develop from the t3 towards the t5 but never reach it; it always adheres the t2. The t4 is widely spaced from the t1; generally, it connects with the t7 only through the enamel and the contact is observable at a medium stage of wear. The t5 is large; it presents good connections with t4 and t6. The t6 is far from the t3; it is the largest cusp of the labial margin of the molar from which it slightly protrudes. The t7 is always present; the contact with the t8 is observable at a medium stage of wear. The t9 is well developed; it is weakly inclined forwards, adhering to t6. The contact with the t8 is generally present due to more or less developed crests. The t8 is well developed, conical and inclined posteriorly. The t12 is

Table 3 Material and measurements of Apodemus gudrunae from Verduno. Dental elements

Length

Width

TNr

mNr

min

mean

max

s

mNr

min

mean

max

s

1

M2

1

1

–

1.53

–

–

1

–

1.41

–

–

2

M1 M2

1 1

1 1

– –

2.15 1.52

– –

– –

1 1

– –

1.40 1.35

– –

– –

5

M1 M2 m1 m2

2 5 2 4

2 4 2 4

1.97 1.31 2.06 1.31

2.03 1.39 2.07 1.36

2.08 1.53 2.08 1.39

– 0.1014 – 0.0331

2 5 2 4

1.27 1.24 1.21 1.15

1.29 1.30 1.23 1.20

1.31 1.37 1.24 1.22

– 0.0563 – 0.0310

Layer

Layer: stratigraphic position of the remains (see Fig. 3); TNr: total number of identified specimens for each dental element; mNr: number of measurable specimens (length or width) for each dental element; s: standard deviation.

S. Colombero et al. / Geobios 46 (2013) 111–125

always present and well-formed; it seems to be a labial extension of t9 with which it is always in contact. There are three roots. M2: The anterior border of the molar is wider than the posterior. The t1 is large; it does not contact the t5; it is in advanced position, slightly exceeding the anterior margin of the t5. The t3 is small; in rare cases it is larger, but never exceeds half of the size of t1. The t5 is well developed, inclined backwards. The connection with the t6 is better developed than that with t4. The t6 is well developed; it is weakly connected through the enamel to t9. The t7 is always present; generally, it has a longitudinally elongated section being of smaller dimensions with respect to the t4; it is connected to t8. The contact with t4 is observable at a medium or even lower stage of wear. The t9 is slightly smaller than t6; it is connected to the t8 by a thin crest. The t8 is large, slightly larger than the t5. The t12 develops from the labial side of the t8; it is generally less developed than in the M1; it is elongated and curved, following the posterolabial margin of the molar. There are three roots. m1: The tma is always present, from medium to large size, but it is always smaller than the cusps of the anteroconid complex. The contact between the two cusps of the anteroconid complex is observable at a medium stage of wear. The complex is asymmetric because the lingual cusp is in a more advanced position than the labial one. Moreover, the latter is always slightly smaller than the first. Posterior cusps form a bilobed, reniform complex; they are slightly inclined forwards. The metaconid is in a more advanced position with respect to the protoconid. Usually, such a complex is connected to the front through a thin ridge that starts from the metaconid and reaches the lingual cusp of the anteroconid. Nonetheless, in worn specimens the contact occurs on both sides. Hypoconid and entoconid are united to form a reniform complex; they are inclined forwards. There is no longitudinal spur. The entoconid is in more advanced position with respect to the hypoconid. The posterior heel is generally well developed, oval or circular in outline, slightly shifted on the lingual side. The c1 is always present, close to the labial margin of the hypoconid, with which it can merge. The size is usually larger than that of the other cusps of the labial cingulum. The labial cingulum usually exhibits two cusplets of equal size, or three or four of variable size. There are two roots. m2: The molar is roughly rectangular in shape. The anterolabial cusp is well developed; it is in an advanced position with respect to the rest of the anterior border of the molar. Protoconid and metaconid are associated to form a reniform complex. The two cusps are inclined slightly anteriorly. Hypoconid and entoconid are associated into a bilobed complex slightly inclined forwards. This complex is isolated from the anterior cusps. The posterior heel is always present and well formed; it is characterized by an elliptical cross-section. The labial cingulum is always present and usually consists of three cusps. The lowest cusp (c1) is located near the hypoconid; it has a circular section, and is usually smaller than that of m1. The other cusps usually are of small size. There are two roots. Remarks: The material of A. gudrunae from Verduno was directly compared with the teeth of A. gudrunae from the type locality of Valdecebro 3 (Spain, MN13; van de Weerd, 1976). The dimensions of the teeth of A. gudrunae from Verduno are consistent with those of the type population of Valdecebro 3. The morphological characteristics of the two populations are very similar, even if the t7 of the M1 from Verduno is generally slightly more developed. The size of the teeth from Verduno is consistent with those of the other populations of A. gudrunae from the Teruel Basin (van de Weerd, 1976; Adrover et al., 1993), Negratı´n-1 (Minwer-Barakat et al., 2009a), Amasya (southwestern Anatolia; Sarica-Filoreau, 2002), being slightly smaller than those of the population of Castelnou 3 (Aguilar et al., 1991). The teeth of

119

Verduno were also compared with specimens of A. gudrunae from the locality of Villastar. The two populations exhibit a broad correspondence as far as regards the morphological features: the contact of the t4 with the t7 is observable at a medium stage of wear both in the M1 and M2, and the t3 of M1 presents short spurs. The material of A. gudrunae from Verduno was compared with the teeth of the populations of Apodemus cf. A. gudrunae from Brisighella 2, 4 and 6 (De Giuli, 1989). On the basis of the comparisons, no differences were detected, except for the presence of a small t1bis in one M1 from BRS 6. Compared with Apodemus gorafensis Ruiz Bustos, Sese´, Dabrio, ˜ a, 1984 from the type locality of Gorafe A (Ruiz Padial and Pen Bustos et al., 1984), several localities of Granada basin (Castillo et al., 1990; Garcı´a-Alix et al., 2008b) and Teruel (Adrover et al., 1988, 1993), the teeth of A. gudrunae from Verduno are smaller in size. Sarica-Filoreau (2002) argued that the t1 of the M1 is a useful character to distinguish A. gudrunae and A. gorafensis. She noted that this tubercle is more developed in A. gorafensis, forming a marked incision between the t1 and the t2 in the lingual outline of M1. In the M1 of Verduno the t1 does not form a deep incision with the t2, and is generally slightly smaller than the t2. The comparative analysis with the population of A. gorafensis from Purcal 4 reveals that the specimens from Verduno display less developed spurs on the t3. The cheek teeth of A. gudrunae from Verduno are larger than those of Apodemus atavus Heller, 1936 from Gundersheim-4 (Fejfar and Storch, 1990) and Florina-Ptolemais-Servia basin (Hordijk and de Bruijn, 2009), and are close or only slightly larger than the molars of A. atavus from Tollo de Chiclana (Minwer-Barakat et al., 2005), Granada basin (Garcı´a-Alix et al., 2008b) and Monte La Mesa (Marchetti et al., 2000). Nevertheless, there are some morphological differences; in the population from Verduno, the contact between the t7 and the t4 is present in the M1 and in the M2, whereas in A. atavus it is rare or even absent. Moreover, in the M1 of A. atavus the t3 exhibits spurs more developed than those of the M1 of A. gudrunae from Verduno. Apodemus gudrunae was reported in many late Turolian localities of Spain (van de Weerd, 1976; Mein et al., 1990; Adrover et al., 1993; Agustı´ et al., 2006; Minwer-Barakat et al., 2009a), France (Martı´n-Sua´rez and Mein, 1998), and Turkey (SaricaFiloreau, 2002). Apodemus cf. A. gudrunae was also recorded in France from Castelnou 3 (Aguilar et al., 1991). In Greece, some populations of Apodemus from the Strimon Basin were formerly assigned to A. gudrunae (de Bruijn, 1989) and subsequently referred to A. gorafensis (Storch and Dahlmann, 1995; Hordijk and de Bruijn, 2009) because of the average size of the specimens. In Italy, some teeth of Apodemus cf. A. gudrunae, extremely similar to those from Verduno, were documented at Brisighella (De Giuli, 1989). Like P. meini, A. gudrunae is a quite common species at the end of the Messinian in many European localities; the present record from Verduno seems to confirm its widespread distribution in the Mediterranean area. Genus Occitanomys Michaux, 1969 Occitanomys sp. Fig. 4(I) Material and measurements: Table 4. Description: M1: A partially preserved M1 was found at Verduno. It lacks its posterior portion and the t4 and the t9 are only partially preserved. The tooth displays a brachiodont crown. Cusps are voluminous and rounded, especially at the base; cusps of the labial side (t3–t6–t9) are well aligned, forming a rectilinear labial margin. The t1 is large; it is connected to the t5 and the t1bis. The t1bis is formed by a thin and sharp ridge between the t1 and the t2. The t2 is large and stout; it widens considerably at the base, resulting in a rounded anterior profile of the tooth. The t3 is smaller

S. Colombero et al. / Geobios 46 (2013) 111–125

120

Table 4 Material and measurements of Occitanomys sp. from Verduno. Dental elements

Length

Width

TNr

mNr

min

mean

max

s

mNr

min

mean

max

s

1

M2

2

2

1.67

1.73

1.78

–

2

1.55

1.55

1.55

–

2

M1

1

–

–

–

–

–

1

–

1.70

–

–

Layer

Layer: stratigraphic position of the remains (see Fig. 3); TNr: total number of identified specimens for each dental element; mNr: number of measurable specimens (length or width) for each dental element; s: standard deviation.

in size with respect to t1 and t2; it is located in a more advanced position than the t1, and does not contact the t5. Between t2 and t3 a tiny swelling of the enamel occurs at the base of the tooth crown. Although the distal portion of the tooth is partially preserved, it is possible to observe the presence of a complete stephanodont crown connecting t4–t5–t6–t8–t9. The connection between t4 and t8 reaches approximately half the height of the t8. As far as regards the t4, the enamel connections with the t5 and t8 can be observed. The t5 is large, inclined backwards. The t6 is aligned with the t3 but it is widely spaced from it. Regarding the t9 only the connections with the t8 and with the t6 can be observed. The large t8 is partially preserved. The t12 is not preserved. M2: The molars present a slightly elongated and rounded shape. Cusps are massive and rounded. The t1 is characterized by a small size; it is slightly inclined inwards. A low and short spur extends towards the t5. The t1bis is present; it displays the same size of t1 or it is smaller, and merges with the t1 at a medium stage of wear. The size of the t3 is smaller than or equals that of the t1; it is not connected to the t5. The tubercles t4–t5–t6–t9–t8 are united to form a complete stephanodont crown. The t4 is larger than t1; it is located at the level of the t6-t9 connection. The t5 is voluminous. The t6 is well developed, inclined backwards and slightly inwards. The t9 is always present, smaller in size with respect to t6, with which it always stays in contact. The t8 is the most developed tubercle. The t12 is absent. There are three roots. Remarks: Our material from Verduno displays the typical characteristics of the genus Occitanomys: brachiodont teeth; medium degree of stephanodonty, expressed by the stephanodont crown formed by the t4–t5–t6–t9–t8 connection in the M1 and M2; backward position of the t1, close to the t5 in the M1. Moreover, these specimens exhibit morphological characteristics that prevent their attribution to the genus Paraethomys. In particular, the M1 and the M2 display a well formed t1bis and the t6 of the M1 is well aligned with the t3 and the t9, thereby producing a straight outline of the labial border of the molar, whereas the t6 of Paraethomys is usually much larger and extends beyond the t3 and the t9. Our material has a larger size with respect to Occitanomys adroveri (Thaler, 1966). Moreover, O. adroveri exhibits less inflated cusps and more developed longitudinal spurs in the t3 of the M1. The specimens from Verduno are smaller than those of Occitanomys ellenbergeri (Thaler, 1966) from Nimes, Se`te and Serrat d’en Vaquer (Michaux, 1969). We directly compared our specimens with those of O. ellenbergeri from Se`te. Some slight morphological differences can be detected: the cusps of O. ellenbergeri are more inflated; the t3 of the M1 is more reduced and more anteriorly placed; the contact between t6 and t9 seems to be lower in O. ellenbergeri. The size of our specimens is larger than the average size of the teeth of Occitanomys brailloni Michaux, 1969 from the type locality of Layna, but also from other localities such as Se`te (Michaux, 1969), Ptolemais 3 and 1, Kardia, Vorio 3, Vorio 1 (van de Weerd, 1979; Hordijk and de Bruijn, 2009), and Tollo de Chiclana 1B (Minwer-Barakat et al., 2005). Nevertheless, the largest specimens of M1 of O. brailloni from Nimes (Michaux, 1969) shows a width close to that of the single specimen found at Verduno. On the other

hand, the material documented herein falls within the size range of Occitanomys montheleni Aguilar et al., 1986 from Mont-He´le`ne. We also compared the material from Verduno with the population of O. brailloni from Layna and with O. montheleni from Mont-He´le`ne and Serrat d’en Vaquer. These two species could be distinguished mainly by the size, which is larger for O. montheleni. However, the size ranges are close and can partially overlap. There are few morphological differences. Usually, the specimens of O. brailloni show more developed spurs on the t3 of M1 and the t6 is close to t3, this latter tubercle being slightly behind in the M1 of O. brailloni. Moreover, a small number of M2 of O. montheleni exhibits the t1bis. It is worth noting that these characteristics are affected by a certain degree of variability and therefore can be evaluated only on large samples. The material documented herein is too scarce and therefore cannot be confidently identified at the specific level. An indeterminate species of the genus Occitanomys was documented at Brisighella 18 (De Giuli, 1989). However, direct comparative analysis with the material from Brisighella allowed us to exclude a similar species-level assignment. Occitanomys sp. from BRS 18 has more developed spurs on the t3 of the M1 than in our material and the cusps are less inflated. However, further analyses are required to identify the species to which the specimens from Brisighella can be referred. The species Occitanomys montheleni is known from a few Early Pliocene localities of France (Aguilar et al., 1986, 1991), whereas Occitanomys brailloni was reported from some Ruscinian localities of France, Spain and Greece (Michaux, 1969; van de Weerd, 1979; Vasileiadou et al., 2003; Minwer-Barakat et al., 2005; Hordijk and de Bruijn, 2009). Nonetheless, the earliest record of this species is that of Rema Marmara (Strimon Basin, Greece), questionably assigned to MN 12 (de Bruijn, 1989). In conclusion, the tooth from Verduno constitutes one of the rare records of a large species of the genus Occitanomys before the Pliocene. Family CRICETIDAE Rochebrune, 1883 Genus Apocricetus Freudenthal, Mein and Martı´n-Sua´rez, 1998 Apocricetus cf. A. barrierei (Mein and Michaux, 1970) Fig. 5(A–D) Material and measurement: Table 5. Description: M1: The molar is stout. The large anterocone is robust; it is slightly inclined posteriorly and bears a weak anterior cingulum. The anterocone is divided into two cusps. A groove is clearly observable, especially on the posterior side, even if a superficial incision is also present on the anterior border. The lingual cusp seems to be slightly more worn than the labial one. The protosinus and anterosinus are closed by the anterolingual cingulum and the anterolabial cingulum, respectively. The double anterolophule is very well developed. The two branches of the anterolophule depart from the two tubercles of the anterocone. The anterior protolophule is absent. The protocone is robust with a pyramidal shape; it forms a very narrow basin with the paracone. The posterior protolophule is well developed. The paracone is of smaller size; it is slightly inclined posteriorly. The mesolophe is absent. The sinus and mesosinus are wide and deep; both are closed by two cingula. The structure of the hypocone is similar to that of the protocone, but slightly smaller in size. The basin located

S. Colombero et al. / Geobios 46 (2013) 111–125

121

Fig. 5. A–D. Apocricetus cf. A. barrierei: A, right M1, V067; B, right M2, V066; C, right m1, V280; D, left m3, V069. E, F. Muscardinus aff. M. vireti: E, left m2, V278; F, left m3, V279. Scale bar: 1 mm.

between the hypocone and metacone is slightly wider than the anterior one. A prominent posteroloph extends from the posterior margin, closing the posterosinus and reaching the metacone in the posterior portion of its base. The metacone is slender and conical in shape, with a circular section. The anterior metalophule extends from the metacone. There is no posterior metalophule. There are four roots. M2: The molar is stout. The anterior border is characterized by the presence of the anterior cingula (anterolophs) that develop along the labial and lingual sides; they curve backwards on both the labial and lingual side, reaching the bases of the paracone and protocone and closing the anterosinus and the protosinus, respectively. The former is deeper and larger than the latter. The anterior protolophule is present. The protocone is large, with a triangular cross-section. It closes a narrow depression with the paracone. The paracone exhibits a smaller size and a conical shape. The mesosinus and the sinus are deep; both are closed by two cingula, the labial one being slightly less developed than the lingual one. The hypocone exhibits a structure comparable to that of the protocone. A well-developed posteroloph extends from the

hypocone towards the metacone, enclosing a deep posterosinus. The metacone is larger than the paracone; it is slightly inclined backwards. All the specimens exhibit a posterior metalophule, which is represented by a short ridge intersecting the posteroloph. m1: The molar is elongate and robust. Cusps are stout. The anteroconid is large; a slight incision can be observed along the posterior wall. The anterolophulid is double in one specimen, while it exhibits only the lingual crest in another specimen; the third specimen is fractured and lacks the anterior portion. A very low labial anterolophid is present, partially enclosing the protosinusid. The lingual anterolophid is absent and the anterosinusid is open. The protoconid is placed in a slightly backward position with respect to the metaconid. A low labial cingulum connects the protoconid to the hypoconid, enclosing the sinusid. A deep basin is enclosed between the protoconid and metaconid. The mesolophid is absent. The hypoconid exhibits the same morphology and proportions as the protoconid; it forms a valley with the entoconid. The posterolophid extends from the hypoconid towards the base of the entoconid, thus enclosing the posterosinusid. The entoconid is slightly larger than the metaconid. The roots are not preserved.

Table 5 Material and measurements of Apocricetus cf. A. barrierei from Verduno. Dental elements Layer 5

M1 M2 m1 m3

Length

Width

TNr

mNr

min

mean

max

s

mNr

min

mean

max

s

2 2 3 1

2 2 3 1

2.66 2.10 2.50 –

2.73 2.15 2.54 2.47

2.80 2.20 2.57 –

– – 0.0496 –

2 2 3 1

1.70 1.70 1.43 –

1.76 1.76 1.46 1.74

1.81 1.81 1.49 –

– – 0.0306 –

Layer: stratigraphic position of the remains (see Fig. 3); TNr: total number of identified specimens for each dental element; mNr: number of measurable specimens (length or width) for each dental element; s: standard deviation.

122

S. Colombero et al. / Geobios 46 (2013) 111–125

m3: The shape of the molar is rectangular and rather elongate. The posterior portion is narrower than the anterior. The labial anterolophid is well developed; it extends from the metaconid along the anterolabial margin of the tooth, curving towards the protoconid and partially enclosing a narrow protosinusid. The lingual anterolophid is absent; however, this may be related to the extreme stage of wear of this specimen. The protoconid is large and forms a narrow valley with the metaconid; it is connected to the anterolophid and to the metaconid through a very short anterolophulid. The metaconid is in an advanced position, smaller in size compared to the protoconid; it is connected to the anterolophid and to the anterolophulid. The protoconid connects directly to the entoconid through a very short appendix. The mesolophid is absent. The sinusid is deep and elongated and extends beyond the longitudinal axis of the molar. A low labial cingulum encloses the sinusid, connecting the protoconid and the hypoconid. The mesosinusid is much smaller and is closed by an extremely low lingual cingulum, which barely contacts the base of the entoconid and the metaconid. The hypoconid is similar in structure and morphology to the protoconid, also showing a similar size. A short anterior appendix extends from the anterolingual margin of this cusp, connecting the entoconid. The posterolophid extends from the hypoconid towards the base of the entoconid, enclosing the posterosinusid. The entoconid is in advanced position. There are two roots, the posterior one being oblique. Remarks: The remains of cricetines from Verduno exhibit the morphological characteristics of the genus Apocricetus Martı´nSua´rez and Freudenthal, 1998, including: absence of mesoloph and mesolophid; absence of a labial spur on the anterolophule of the M1; elongate m3; deep valley between the protoconid and the metaconid of m1; stout and crest-like anteroconid of the m1. We compared the material from Verduno with Apocricetus barrierei (Mein and Michaux, 1970) from the type locality of Chabrier, and from the localities of La Gloria 4 and Caravaca. The dimensions of the specimens from Verduno are very close to those of A. barrierei from Chabrier, except for the m3, which is slightly larger in our material. Moreover, the size of specimens described herein is close to that of A. barrierei from Hautimagne, Vendargues (Mein and Michaux, 1970), Font Estramar (Aguilar et al., 1991), La Gloria 4 (Adrover et al., 1993), Yeguas and Purcal 4 (Garcı´a Alı´x et al., 2008a). The morphology of the specimens from Verduno is very similar to that of A. barrierei from the type locality of Chabrier given that no significant differences can be easily detected. With respect to the population of A. barrierei from Caravaca, the material from Verduno differs for the presence of the anterior protolophule and posterior metalophule in all the M2, whereas in the material from Caravaca, the anterior protolophule may be absent and there is not a posterior metalophule. Nonetheless, an anterior protolophule is always present in the populations of A. barrierei from Purcal 4, and the posterior metalophule is present in the half of the M2 from Almadumena 1D (Freudenthal et al., 1998). The comparative analysis of the material of Apocricetus cf. A. barrierei from Brisighella 3 and 6 (Cricetus cf. C. barrierei in the original diagnosis; De Giuli, 1989) revealed that the M2 of these populations display a posterior metalophule. Compared to Apocricetus alberti Freudenthal, Mein and Martı´nSua´rez, 1998, the remains of Verduno exhibit much larger size, particularly regarding the M1 and m3. The morphological characteristics of A. alberti from the type locality of Crevillente 6 (see de Bruijn et al., 1975; Freudenthal et al., 1998) are quite similar to those of the specimens from Verduno. However, some differences can be easily detected. In particular, an anterior protolophule is often present in the M1 from Crevillente 6, whereas it is always absent in the material from Verduno, and the m3 in some cases exhibit a relatively developed mesolophid (Freudenthal et al., 1998).

A comparison with Apocricetus angustidens (Depe´ret, 1890) from the type locality of Serrat d’en Vaquer indicates that the size and proportions of the specimens from Verduno are smaller except for the m3 that fall within the same size range. Moreover, the from Verduno specimens are smaller than those from other localities such as Mont-He´le`ne (Aguilar et al., 1986; Freudenthal et al., 1998), Le Soler, Villeneuve la Raho and Port-La-Nouvelle (Freudenthal et al., 1998). The specimens from Verduno differ from those of A. angustidens from Serrat d’en Vaquer for the less developed anterior cingulum of the M1 and the presence of the posterior metalophule in the M2. Finally, the material from Verduno differs from A. angustidens from Mont-He´le`ne since the M2 of this population exhibits an anterior protolophule and a posterior metalophule in very few specimens (Freudenthal et al., 1998). A. barrierei could be regarded as an intermediate form between A. alberti and A. angustidens. This phyletic lineage (A. alberti – A. barrierei – A. angustidens) is characterized by an increasing in size and by the gradual loss or reduction of the anterior protolophule and posterior metalophule in the upper molars (Freudenthal et al., 1998). The cricetines from Verduno are assigned to Apocricetus cf. A. barrierei as they fall in the size range of A. barrierei, although displaying some plesiomorphic features such as the presence of the posterior metalophule and the anterior protolophule in the M2. These characteristics are also present in the population of Apocricetus cf. A. barrierei from Brisighella 3 and 6. In Italy, Apocricetus cf. A. barrierei is also reported at Borro Strolla (probably close to the MN13-MN14 limit; Abbazzi et al., 2008). As far as we known the Italian findings of Apocricetus cf. A. barrierei may represent the easternmost record of the genus. Family GLIRIDAE Thomas, 1897 Genus Muscardinus Kaup, 1829 Muscardinus aff. M. vireti Hugueney and Mein, 1965 Fig. 5(E, F) Material and measurements: 1 m2 (L  W: 1.40  1.29 mm), 1 m3 (1.16  1.23 mm). Description: m2: The molar is subsquared. The lingual border is slightly longer than the labial one. There are six ridges generally convex anteriorly and slightly inclined forward. The labial side of the first crest is in a retracted position with respect to the lingual side. On the contrary, in the other ridges the lingual side is in the rearmost position. The first and the second ridges are connected along the labial side. The anterolabial corner is more rounded than the lingual one. Between the third and fourth ridges, the labial and lingual portions of an extra-ridge can be observed. The lingual portion is more developed. Roots are not preserved. m3: The molar exhibits a subtrapezoidal outline, with the posterior border shorter than the anterior one. The labial border is shorter than the lingual one since the distance between the ridges on the labial side is shorter than that on the lingual side. There are six ridges anteriorly convex except for the two posterior ones, which are oblique and straight. The first and second ridges, and fifth and sixth ridges are connected along the labial side. A small portion of an extra-ridge occurs on the lingual side between the third and fourth ridges. No roots are preserved. Remarks: The teeth of Muscardinus from Verduno were compared to those of Muscardinus vireti Hugueney and Mein, 1965 from the type locality of Lissieu. The specimens from Verduno are slightly larger, showing similar morphological characters. The single m2 collected from Verduno exhibit six complete ridges and lingual and labial portions of a seventh extra-ridge are present between the 3rd and 4th ridges. This pattern can be observed approximately in the half of the specimens from Lissieu. The m3 from Lissieu slightly differs from those of the material from Verduno by the absence of any extra-ridge.

S. Colombero et al. / Geobios 46 (2013) 111–125

The teeth of Muscardinus aff. M. vireti from Baccinello V3 (Engesser, 1983) are slightly smaller than those from Verduno, but they display a very similar morphological pattern. Indeed, as in the specimens from Verduno, the m2s and m3s from Baccinello V3 exhibit an incomplete extra-ridge between the 3rd and 4th ridges. The material from Verduno can be easily separated from Muscardinus dacicus Kormos, 1930 for their smaller size. Moreover, the morphological pattern of m2 and m3 of M. dacicus is different, not displaying any extra-ridges (Daoud, 1993). The specimens from Verduno are slightly larger than Muscardinus pliocaenicus Kowalski, 1963 from the type locality of We˛z´e (Kowalski, 1963), but also from other localities such as Re˛bielice, Podlesice (Daoud, 1993), Tegelen and Zuurland (Reumer, 2001; van den Hoek Ostende, 2003). The teeth of Muscardinus pliocaenicus austriacus Bachmayer and Wilson, 1970 from the localities of Kohfidisch, Eichkogel and Schernham, usually are slightly smaller than the specimens from Verduno (Daxner-Ho¨ck and Ho¨ck, 2009). The material from Verduno also differs from these two taxa for the presence of an extra-ridge on the m2 and the m3. The size of Muscardinus topachevskii Nesin and Kowalski, 1997 is smaller than that of our specimens. Moreover, the morphological pattern of M. topachevskii is simpler, not displaying the extraridges between the 3rd and 4th ridges on the m2. The size and proportions of the teeth of Muscardinus meridionalis Garcı´a-Alix, Minwer-Barakat, Martı´n-Sua´rez and Freudenthal, 2008 are close to those of our specimens. The m2s and m3s of M. meridionalis display a small lingual portion of an extra-ridge in a very few specimens (Garcı´a-Alix et al., 2008c). On the contrary, the lingual portion of the extra-ridge is always present in the specimens from Verduno. Moreover the m2 from Verduno exhibits also the labial portion of the extra-ridge, which is absent in the m2 of M. meridionalis. However, it should be noted that the material from Verduno is very sparse and the variability cannot be properly evaluated. Muscardinus helleri Fejfar and Storch, 1990 from Gundersheim 4 presents a similar morphological pattern and size compared to the specimens from Verduno. However, the m2 of M. helleri is characterized by a larger width. Nevertheless, the pattern of the m3 of M. helleri is unknown and the sample size of the population from Gundersheim 4 is too small to provide a reliable comparison. In conclusion, the general pattern of the fossil teeth of Muscardinus from Verduno is somewhat close to that of M. vireti from Lissieu, and particularly similar to Muscardinus aff. M. vireti from Baccinello V3. Therefore, we prefer to assign the material from Verduno to Muscardinus aff. M. vireti because of its larger size and for the presence of an extra-ridge in the m3. 6. Discussion The fossil rodents from Verduno presented herein provide new significant information about the terrestrial vertebrate communities of a poorly known area of the northern sector of the Mediterranean region. The rodent assemblage from Verduno brings the number of late Messinian continental vertebratebearing localities from Piedmont to three, the two others being Moncucco Torinese (Angelone et al., 2011) and Ciabo`t Cagna (Cavallo et al., 1993). Despite some similarities, the three fossil assemblages exhibit structural differences, providing evidence of the complex environmental scenario that characterized the terrestrial ecosystems of the TPB at the end of the Miocene.

123

taxonomic composition, with a very similar relative abundance of the different taxa, exemplified by the sharp dominance of C. benericettii and P. meini (Tables 1 and 2). On the contrary, layer 5 displays a slightly different composition, with the exclusive presence of Apocricetus cf. A. barrierei and Muscardinus aff. M. vireti, and the common presence of Apodemus gudrunae. It is not possible to definitely exclude that the non-homogeneous taxonomic composition throughout the exploited productive layers might be the consequence of non-exhaustive sampling. Indeed, some taxa, such as the cricetines, are normally quite rare with respect to other rodents. 6.2. Depositional environments The sedimentary history of the continental deposits characterizing the upper portion of the Messinian section of Verduno has not been investigated in detail. However, as far as the excavation site is concerned, layer 5 may be interpreted as the product of fluvial deposition, as suggested by the trough cross-bedded gravels of the channelized conglomerates, which are separated from the lowest layers by a clear erosive surface. Mollusks and ostracods are currently under study and will certainly provide a better definition of the depositional environment of each layer. Fossils from upper conglomerates (layer 5) are clearly reworked, while the discovery of partially articulated skeletal remains of small and large mammals in the partially laminated silty clays of layer 1 seems to indicate a null or very low degree of resedimentation. 6.3. Biochronological remarks As discussed above, the vertebrate-bearing deposits of Verduno are included in the Cassano Spinola Conglomerates Fm., a unit, which can be correlated with the post-evaporitic phase of the Messinian Salinity Crisis (5.55–5.33 Ma; Dela Pierre et al., 2007, 2011; CIESM, 2008). Most of the fossil rodents recognized in the Verduno assemblages concur to indicate a post-evaporitic Messinian age, which is therefore consistent with the stratigraphic interpretation. However, some comments about the distribution and morphology of some recognized taxa are necessary in order to properly interpret their stratigraphic significance. The large cricetine Apocricetus cf. A. barrierei is relatively common at the beginning of the Ruscinian (Freudenthal et al., 1998; MinwerBarakat, 2005); however, as discussed above, the material from Verduno exhibits some plesiomorphic characters, including the presence of the posterior metaloph in the M2, which may be easily explained by an older stratigraphic position. The slight morphological differences, evidenced in this study, between the material from Verduno referred to C. benericettii and that from other Messinian post-evaporitic localities (Brisighella and Moncucco Torinese; De Giuli, 1989; Angelone et al., 2011) are difficult to interpret. Those differences may be related to different paleoecological constraints rather than to stratigraphic differences. Finally, the different taxonomic composition of Verduno with respect to the other, approximately correlated rodent assemblages from Piedmont, i.e., the presence of Gerbillidae gen. and sp. indet. at Ciabo`t Cagna, and the great abundance of glirids at Moncucco Torinese, cannot be considered as related to a different age. Therefore, based on the available data furnished by the fossil rodents composition, we assign the vertebrate assemblage from Verduno to the terminal part of the MN13, very close to the Turolian/Ruscinian boundary.

6.1. Composition of the rodent assemblages 6.4. Palaeoecological considerations The systematic analysis of the rodent assemblages has revealed a relatively conservative taxonomic composition throughout the fossiliferous layers. Layers 1 and 2 show a nearly identical

On the grounds of the recovered small mammal fossils from Verduno, it is possible to make some general paleoecological

124

S. Colombero et al. / Geobios 46 (2013) 111–125

considerations, especially on layers 1 and 2. As a matter of fact, we are dealing with a poor and unbalanced association: indeed, up to date, among rodents only murids have been found in these layers. Glirids are totally absent in these two basal layers, being very rare in the overlying deposits (layer 5). Among murids, Apodemus gudrunae and Occitanomys sp. are very rare, whereas the assemblages are clearly dominated by the small C. benericettii and the large P. meini murids. According to Martı´n-Sua´rez et al. (2001), Apodemus gudrunae can be considered as an index of warm and wet conditions, whereas P. meini indicates warm and dry conditions though this hypothesis cannot be confirmed by the presence of C. benericettii and Occitanomys sp., whose climatic and palaeoecological preferences are still unknown. The absence of glirids, the scarcity of the genus Apodemus, and the sharp dominance of taxa like P. meini concur to indicate the dominance of open habitats to the possible detriment of arboreal ecological niches. Besides rodents, the other vertebrate taxa are currently under study and will be the object of further works. They will certainly provide useful data to properly understand the overall palaeoecological context. 6.5. Palaeobiogeographic considerations As extensively discussed by Rook et al. (2006), the Italian continental vertebrate localities document the existence of three well-defined Late Miocene paleobioprovinces, including the endemic Apulo-Abruzzo bioprovince, the Tusco-Sardinia bioprovince, and the Calabria-Sicily bioprovince of North African affinities. The paleobiogeographic character of Northern Italian land-vertebrate localities have been never discussed in detail and properly interpreted. As briefly argued above, the biogeographic characteristics of the Verduno rodent assemblages are clearly defined through a direct faunal comparison with those from other, apparently coeval Northern Italian localities such as Brisighella (Central Italy; De Giuli, 1989), Ciabo`t Cagna (NW Italy; Cavallo et al., 1993), and Moncucco Torinese (NW Italy; Angelone et al., 2011). The small mammal assemblage from Brisighella displays a high taxonomic diversity, with glirids (Muscardinus sp., Myomimus sp.) and sciurids (Atlantoxerus cf. A. rhodius, Hylopetes sp.); Stephanomys debruijni, which is absent at Verduno, is by far the most common taxon. The small mammal assemblage from Ciabo`t Cagna is poorly diversified, even though the presence of a member of the family Gerbillidae clearly evidences its peculiar taxonomic composition, suggesting a sharp separation from Verduno. As far as the rodent assemblage from Moncucco Torinese is concerned (Angelone et al., 2011), it appears to be characterized by a relatively diverse glirid content (Muscardinus cf. M. vireti, Glis minor, Eliomys cf. E. truci) and by a population of C. benericettii with a slightly minor degree of stephanodonty when compared with Verduno. These differences possibly reflect a different paleoenvironmental setting, as suggested by the presence of Tapirus sp. and the glirids (Angelone et al., 2011), which are indicative of the presence of an extensive forest coverage. Overall, considered from a biogeographic point of view, the rodent assemblage from Verduno consists of species with ubiquitous affinity such as P. meini and Apodemus gudrunae, associated with a species of western origin (Apocricetus cf. A. barrierei). 7. Conclusions The fossil rodents from Verduno provide a variety of relevant information concerning this recently-discovered latest Messinian locality of Northern Italy. From a biochronological point of view, the rodent assemblage can be referred to the MN13 unit. These data agree with the age of the deposits that are referable to the post-evaporitic phase of the Messinian Salinity Crisis

(5.55–5.33 Ma; CIESM, 2008), close to the Mio-Pliocene boundary. From a paleoecological point of view, the fossil rodents from Verduno point to the possible dominance of open spaces and the consequent restricted presence of closed habitats. Nonetheless, data from the other vertebrate taxa are necessary in order to provide an exhaustive analysis of the paleoecological context. Due to its geographic position, the locality of Verduno appears to be crucial to interpret the general biogeographic characters of the northern Italian terrestrial paleobiotopes during the late Messinian, and to properly define the immigration pathways toward the emerging Apennine chain from both the western and eastern sectors of the European continent. Acknowledgements This study was financially supported by University Ministry grants MIUR PRIN 2009MSSS9L-002 (resp. G. Pavia, Torino). Federica Giudice improved the English. Dr. Simona Cavagna operated the SEM of the Dipartimento di Scienze della Terra, Universita` degli Studi di Torino. We thank the reviewers Drs. Raef Minwer-Barakat, Carmen Sese´ and Pierre Mein for the useful suggestions that definitely improved the manuscript. We are grateful to Drs. Marguerite Hugueney, Pierre Mein and Abel Prieur of the University of Lyon 1, who kindly allowed us to visit their fossil rodent collections. We would also thank Drs. Wilma Wessels and Hans de Bruijn of the IVAU (Utrecht) for their hospitality and kindness during our visit to their fossil rodent collections. We are grateful to Prof. Lorenzo Rook of the Universita` degli Studi di Firenze for his helpfulness during our visit to the fossil rodent collection from Brisighella. This work would not have been possible without the collaboration of the students of the Universita` degli Studi di Torino and volunteers of the Museo Civico ‘‘Federico Eusebio’’ (Alba) during the excavation campaigns. In particular, we greatly appreciated the helpful collaboration of Chiara Angelone, Cristina Bagnus, Oreste Cavallo, Piero Giuntelli, Boris Villier and Marta Zunino. References Abbazzi, L., Benvenuti, M., Ceci, M.E., Esu, D., Faranda, C., Rook, L., Tangocci, F., 2008. The end of the Lago-Mare time in the SE Valdelsa Basin (Central Italy): interference between local tectonism and regional sea-level rise. Geodiversitas 30, 611–639. Adrover, R., Mein, P., Moissenet, E., 1988. Contribucio´n al conocimiento de la fauna de roedores del Plioceno de la regio´n de Teruel. Teruel 79, 91–151. Adrover, R., Mein, P., Moissenet, E., 1993. Roedores de la transicio´n Mio-Plioceno de la regio´n de Teruel. Paleontologia i Evolucio 26/27, 47–84. Aguilar, J.P., Calvet, M., Michaux, J., 1986. Description des rongeurs plioce`nes de la faune du Mont-He´le`ne (Pyre´ne´es orientales, France), nouveau jalon entre les faunes de Perpignan (Serrat-d’en-Vacquer) et de Se`te. Palaeovertebrata 16, 127–144. Aguilar, J.P., Clauzon, G., Michaux, J., 1989. La limite Mio-Plioce`ne dans le Sud de la France d’apre`s les faunes de rongeurs : e´tat de la question et remarques sur les datations a` l’aide des rongeurs. Bollettino della Societa` Paleontologica Italiana 28, 137–145. Aguilar, J.P., Escarguel, G., Michaux, J., 1999. A succession of Miocene rodent assemblages from fissure fillings in southern France: palaeoenvironmental interpretation and comparison with Spain. Palaeogeography Palaeoclimatology Palaeoecology 145, 215–230. Aguilar, J.P., Michaux, J., Bachelet, B., Calvet, M., Faillat, J.P., 1991. Les nouvelles faunes de rongeurs proches de la limite Mio-Plioce`ne en Roussillon. Implications biostratigraphiques et bioge´ographiques. Palaeovertebrata 20, 147–174. Agustı´, J., Garce´s, M., Krijgsman, W., 2006. Evidence for African-Iberian exchanges during the Messinian in the Spanish mammalian record. Palaeogeography Palaeoclimatology Palaeoecology 238, 5–14. Agustı´, J., Llenas, M., 1996. The late Turolian muroid rodent succession in eastern Spain. Acta Zoologica Cracoviensia 39, 47–56. Angelone, C., Colombero, S., Esu, D., Giuntelli, P., Marcolini, F., Pavia, M., Trenkwalder, S., van den Hoek Ostende, L., Zunino, M., Pavia, G., 2011. Moncucco Torinese, a new post-evaporitic Messinian fossiliferous site from Piedmont (NW Italy). Neues Jahrbuch fu¨r Geologie und Pala¨ontologie Abhandlungen 259, 89–104. Bernardi, E., Dela Pierre, F., Irace, A., Lozar, F., Natalicchio, M., Clari, P., Martire, L., Violanti, D., 2010. The Messinian sediments of the Tertiary Piedmont Bas new data from outcrop sections. GEOSED 2010, Post congress field trip, 24–25 September, pp. 35.

S. Colombero et al. / Geobios 46 (2013) 111–125 Bertini, A., Martinetto, E., 2011. Reconstruction of vegetation transects for the Messinian-Piacenzian of Italy by means of comparative analysis of pollen, leaf and carpological records. Palaeogeography Palaeoclimatology Palaeoecology 304, 230–246. Boni, A., Casnedi, R., 1970. Note Illustrative Della Carta Geologica D’Italia Alla Scala 1:100.000, Fogli 69–70 Asti e Alessandria. Ministero dell’industria, del commercio e dell’artigianato, Direzione generale delle miniere, Servizio geologico d’Italia, Italia, pp. 64. de Bruijn, H., 1989. Smaller mammals from the Upper Miocene and Lower Pliocene of the Strimon basin, Greece. Part 1. Rodentia and Lagomorpha. Bollettino della Societa` Paleontologica Italiana 28, 189–195. de Bruijn, H., Dawson, M.R., Mein, P., 1970. Upper Pliocene Rodentia, Lagomorpha and Insectivora (Mammalia) from the Isle of Rhodes (Greece). I–III. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen B 73, 535–584. de Bruijn, H., Mein, P., Montenat, C., van de Weerd, A., 1975. Corre´lations entre les gisements de rongeurs et les formations marines du Mioce`ne terminal d’Espagne me´ridionale I (Provinces d’Alicante et de Murcia). Proceedings Koninklijke Nederlandse Akademie van Wetenschappen B, 78, 1–32. Castillo, C., Freudenthal, M., Martı´n-Sua´rez, E., Martinez, M.V., Rivas, P., 1990. New localities with fossil micromammals in the Pliocene of the Granada Basin (Spain). Scripta Geologica 93, 41–46. Cavallo, O., Sen, S., Rage, J.C., Gaudant, J., 1993. Verte´bre´s messiniens du facie`s a` Conge´ries de Ciabo`t Cagna, Corneliano d’Alba (Pie´mont, Italie). Rivista Piemontese di Storia Naturale 14, 3–22. CIESM, 2008. The Messinian Salinity Crisis from mega-deposits to microbiology – A consensus report. CIESM Workshop Monographs 33, 168 pp. Clari, P., Bernardi, E., Cavagna, S., Dela Pierre, F., Irace, A., Lozar, F., Martinetto, E., Trenkwalder, S., Violanti, D., 2008. Alba e tramonto della crisi messiniana, Alba 10–11 Ottobre 2008, Dipartimento di Scienza della Terra, Torino. Dam, J.A., van, Alcala´, L., Zarza, A.A., Calvo, J.P., Garce´s, M., Krijgsman, W., 2001. The upper Miocene mammal record from the Teruel-Alfambra region (Spain). The MN system and continental stage/age concepts discussed. Journal of Vertebrate Paleontology 21, 367–385. Daoud, A., 1993. Evolution of Gliridae (Mammalia, Rodentia) in Pliocene and Quaternary of Poland. Acta Zoologica Cracoviensa 36, 199–231. De Giuli, C., 1989. The rodents of the Brisighella latest Miocene Fauna. Bollettino della Societa` Paleontologica Italiana 28, 197–212. Dela Pierre, F., Bernardi, E., Cavagna, S., Clari, P., Gennari, R., Irace, A., Lozar, F., Lugli, S., Manzi, V., Natalicchio, M., Roveri, M., Violanti, D., 2011. The record of the Messinian salinity crisis in the Tertiary Piedmont Basin (NW Italy): the Alba section revisited. Palaeogeography Palaeoclimatology Palaeoecology 310, 238–255. Dela Pierre, F., Festa, A., Irace, A., 2007. Interaction of tectonic, sedimentary, and diapiric processes in the origin of chaotic sediments: An example from the Messinian of Torino Hill (Tertiary Piedmont Basin, northwestern Italy). Bulletin of the Geological Society of America 119, 1107–1119. Daxner-Ho¨ck, G., Ho¨ck, E., 2009. New data on Eomyidae and Gliridae (Rodentia, Mammalia) from the Late Miocene of Austria. Annalen des Naturhistorischen Museums in Wien 111A, 375–444. Einsele, G., 1992. Sedimentary Basins: Evolution, Facies and Sediment Budget. Springer-Verlag, Berlin. Engesser, B., 1983. Die jungtertia¨ren Kleinsa¨uger des Gebietes der Maremma (Toskana, Italien). 1. Teil: Gliridae (Rodentia, Mammalia). Eclogae Geologicae Helvetiae 76, 763–780. Fejfar, O., Storch, G., 1990. Eine Plioza¨ne (ober-Ruscinische) Kleinsa¨ugerfauna aus Gundersheim, Rheinhessen-1. Nagetiere: Mammalia, Rodentia. Senckenbergiana Lethaea 71, 139–184. Freudenthal, M., Martı´n-Sua´rez, E., 1999. Family Muridae. In: Ro¨ssner, G., Heissig, K. (Eds.), The Miocene Land Mammals of Europe. Verlag Dr. Friedrich Pfeil, Mu¨nchen, pp. 401–409. Freudenthal, M., Mein, P., Martı´n-Sua´rez, E., 1998. Revision of Late Miocene and Pliocene Cricetinae (Rodentia, Mammalia) from Spain and France. Treballs del Museu de Geologia de Barcelona 7, 11–93. Garce´s, M., Krijgsman, W., Agustı´, J., 1998. Chronology of the late Turolian deposits of the Fortuna basin (SE Spain): implications for the Messinian evolution of the eastern Betics. Earth and Planetary Science Letters 163, 69–81. Garcı´a-Alix, A., Minwer-Barakat, R., Martı´n-Sua´rez, E., Freudenthal, M., 2008a. Cricetidae and Gliridae (Rodentia, Mammalia) from the Miocene and Pliocene of southern Spain. Scripta Geologica 136, 1–37. Garcı´a-Alix, A., Minwer-Barakat, R., Martı´n-Sua´rez, E., Freudenthal, M., 2008b. Muridae (Rodentia, Mammalia) from the Mio-Pliocene boundary in the Granada Basin (southern Spain). Biostratigraphic and phylogenetic implications. Neues Jahrbuch fu¨r Geologie und Pala¨ontologie Abhandlungen 248 (2), 183–215. Garcı´a-Alix, A., Minwer-Barakat, R., Martı´n-Sua´rez, E., Freudenthal, M., 2008c. Muscardinus meridionalis sp. nov., a new species of Gliridae (Rodentia, Mammalia) and its implications for the phylogeny of Muscardinus. Journal of Vertebrate Paleontology 28, 568–573. Garcı´a-Alix, A., Minwer-Barakat, R., Martin, J., Martı´n-Sua´rez, E., Freudenthal, M., 2008d. Biostratigraphy and sedimentary evolution of Late Miocene and Pliocene continental deposits of the Granada Basin (southern Spain). Lethaia 41 (4), 431–446. Ghibaudo, G., Clari, P., Perello, M., 1985. Litostratigrafia, sedimentologia ed evoluzione tettonico-sedimentaria dei depositi miocenici del margine sud-orientale del Bacino Terziario Ligure-Piemontese (Valli Borbera, Scrivia e Lemme). Bollettino della Societa` Geologica Italiana 104, 349–397.

125

Hoek Ostende, L. van den., 2003. Gliridae (Rodentia, Mammalia) from the Upper Pliocene of Tegelen (province of Limburg, The Netherlands). Scripta Geologica 126, 203–215. Hordijk, K., de Bruijn, H., 2009. The succession of rodent faunas from the Mio/ Pliocene lacustrine deposits of the Florina-Ptolemais-Servia Basin (Greece). Hellenic Journal of Geosciences 44, 21–103. Hugueney, M., Mein, P., 1965. Lagomorphes et rongeurs du Ne´oge`ne de Lissieu (Rhoˆne). Travaux du Laboratoire de Ge´ologie de la Faculte´ des Sciences de Lyon Nouvelle Serie 12, 109–123. Kowalski, K., 1963. The Pliocene and Pleistocene Gliridae (Mammalia, Rodentia) from Poland. Acta Zoologica Cracoviensia 8, 533–567. Marchetti, M., Parolin, K., Sala, B., 2000. The Biharian fauna from Monte La Mesa (Verona, northeastern Italy). Acta Zoologica Cracoviensia 43, 79–105. Martı´n-Sua´rez, E., Freudenthal, M., 1998. Biostratigraphy of the continental Upper Miocene of Crevillente (Alicante, SE Spain). Geobios 31, 839–847. Martı´n-Sua´rez, E., Freudenthal, M., Civis, J., 2001. Rodent palaeoecology of the Continental Upper Miocene of Crevillente (Alicante, SE Spain). Palaeogeography Palaeoclimatology Palaeoecology 165, 349–356. Martı´n-Sua´rez, E., Mein, P., 1991. Revision of the genus Castillomys (Muridae, Rodentia). Scripta Geologica 96, 47–81. Martı´n-Sua´rez, E., Mein, P., 1998. Revision of the Genera Parapodemus, Apodemus, Rhagamys and Rhagapodemus (Rodentia, Mammalia). Geobios 31, 87–97. Mein, P., Michaux, J., 1970. Un nouveau stade dans l’e´volution des rongeurs plioce`nes de l’Europe sud-occidentale. Comptes Rendus de l’Acade´mie des sciences de Paris (D) 270, 2780–2783. Mein, P., Moissenet, E., Adrover, R., 1990. Biostratigraphie du Ne´oge`ne supe´rieur du bassin de Teruel. Paleontologia i Evolucio´ 23, 121–139. Michaux, J., 1969. Muridae (Rodentia) du Plioce`ne supe´rieur d’Espagne et du Midi de la France. Palaeovertebrata 3, 1–25. Minwer-Barakat, R., 2005. Roedores e insectı´voros del Turoliense superior y el Plioceno del sector central de la cuenca de Guadix. Ph.D Thesis. Departamento de Estratigrafı´a y Paleontologı´a, Granada. Minwer-Barakat, R., Garcı´a-Alix, A., Agustı´, J., Martı´n-Sua´rez, E., Freudenthal, M., 2009a. The micromammal fauna from Negratin-1 (Guadix Basin, Southern Spain): new evidence of African-Iberian mammal exchanges during the late Miocene. Journal of Paleontology 83, 854–879. Minwer-Barakat, R., Garcı´a-Alix, A., Martı´n-Sua´rez, E., Freudenthal, M., 2005. Muridae (Rodentia) from the Pliocene of Tollo de Chiclana (Granada, Southern Spain). Journal of Vertebrate Paleontology 25, 426–441. Minwer-Barakat, R., Garcı´a-Alix, A., Martı´n-Suarez, E., Freudenthal, M., 2009b. Late Turolian micromammals from Rambla de Chimeneas-3: considerations on the oldest continental faunas from the Guadix Basin (Southern Spain). Neues Jahrbuch fu¨r Geologie und Pala¨ontologie Abhandlungen 83 (6), 854–879. Montenat, C., de Bruijn, H., 1976. The Ruscinian rodent faunule from La Juliana (Murcia); its implication for the correlation of continental and marine biozones. Proceedings Koninklijke Nederlandse Akademie Van Wetenschappen 79, 245–255. Reumer, J.W.F., 2001. Gliridae (Mammalia, Rodentia) from the Zuurland boreholes near Rotterdam, the Netherlands. Deinsea 8, 41–48. Rook, L., Gallai, G., Torre, D., 2006. Lands and endemic mammals in the Late Miocene of Italy: constraints for paleogeographic outlines of Tyrrhenian area. Palaeogeography Palaeoclimatology Palaeoecology 238, 263–269. Roveri, M., Manzi, V., 2007. Gessoso-Solfifera. In: Cita, M.B., Abbate, E., Aldighieri, B., Balini, M., Conti, M.A., Falorni, P., Germani, D., Groppelli, G., Manetti, P., Petti, F.M. (Eds.), Catalogo Delle Formazioni. Unita` Tradizionali. Carta Geologica D’Italia 1:50. 000. Serie III 7. Quaderni del Servizio Geologico d’Italia, Italia, pp. 303–310. ˜ a, J.A., Padial, J., 1984. Geologia y fauna de Ruiz Bustos, A., Sese´, C., Dabrio, C., Pen micromamiferos del nuevo yacimiento del Plioceno Inferior de Gorafe A (depresion de Guadix-Baza, Granada). Estudios geologicos 40, 231–241. Sardella, R., 2008. Remarks on the Messinian carnivores (Mammalia) of Italy. Bollettino della Societa Paleontologica Italiana 47, 195–202. Sarica-Filoreau, N., 2002. Faunes de rongeurs ne´oge`nes et quaternaires des grabens d’Anatolie occidentale. Syste´matique, biochronologie et implications tectoniques. Ph.D. Thesis, Muse´um National d’Histoire Naturelle, Paris. Sen, S., Jaeger, J.J., Dalfes, N., Mazin, J.M., Bocherens, H., 1989. Discovery of a small mammal fauna in western Anatolia Pliocene. Comptes Rendus de l’Acade´mie des Sciences de Paris (2) 309, 1729–1734. Storch, G., Dahlmann, T., 1995. The vertebrate locality Maramena (Macedonia, Greece) at the Turolian-Ruscinian boundary (Neogene). 10. Murinae (Rodentia, Mammalia). Mu¨nchner Geowissenschaftliche Abhandlungen 28, 121–132. Sturani, C., 1973. A fossil eel (Anguilla sp.) from the Messinian of Alba (Tertiary piedmontese basin). Palaeoenvironmental and palaeogeographic implications. Messinian events in the Mediterranean. Koninklijke Nederlandse Academie van Wetenschappen 243–255. Sturani, C., 1976. Messinian facies in the Piedmont basin. Memorie della Societa` Geologica Italiana 16, 11–25. Vasileiadou, K.V., Koufos, G.D., Syrides, G.E., 2003. Silata, a new locality with micromammals from the Miocene/Pliocene boundary of the Chalkidiki peninsula, Macedonia, Greece. Deinsea 10, 549–562. van de Weerd, A., 1976. Rodent faunas of the Mio-Pliocene continental sediments of the Teruel-Alfambra region, Spain. Utrecht Micropaleontological Bulletins 2, 1– 217. van de Weerd, A., 1979. Early Ruscinian rodents and lagomorphs (Mammalia) from the lignites near Ptolemais (Macedonia, Greece). Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen B 82, 127–170.