Introduction

Southeastern Europe has always been regarded as an important paleogeographic area because of its crucial role on the mammal exchanges between Asia, Africa, and Europe (Pickford and Morales 1994; Rögl 1999; Koufos et al. 2005; Wessels 2009; Álvarez-Sierra et al. 2013; De Bruijn et al. 2013). Asia Minor has been an important source of European micromammals during the Early Miocene and especially for Muroidea rodents. In this way, there are abundant genera of Miocene cricetids, such as Cricetodon, Democricetodon, Karydomys, and Megacricetodon, that occurred in Anatolia much earlier than in Europe (Klein Hofmeijer and De Bruijn 1988; Theocharopoulos 2000; Doukas 2003). Even though results on insectivores and eomyids may indicate a connection between Greece and western Europe around MN 3 (Van den Hoek Ostende 2001), the presence in Greece of the earliest representatives of many of these genera in Europe supports the generally accepted idea that this area was connected to Asia Minor during the Early Miocene but that the connection with central and western Europe, probably through the Dinarids was not possible till Middle Miocene (Popov et al. 2004; Koufos et al. 2005; Wessels 2009). While the earlier Cricetodontini and Democricetodontinae from Europe have been studied from a paleogeographical point of view (De Bruijn et al. 1993; Theocharopoulos 2000; Álvarez-Sierra et al. 2013) much less is known about the genus Megacricetodon, despite it being one of the most diverse and abundant rodent during the Early and Middle Miocene in Europe.

The goal of this work is the study of the earliest European Megacricetodon material from the Greek locality of Aliveri (North and South quarries). Based on this study, and the new results and interpretations proposed by Oliver and Peláez-Campomanes (2013, 2014) on the early representatives of the central and western European record of the genus Megacricetodon, a palaeobiogeographical framework for the evolution of the group is proposed.

The site of Aliveri (Island of Evia, Greece) is situated in the Neogene sedimentary basin of Aliveri-Kymi. It was discovered by Hans De Bruijn, Albert J. Van der Meulen, and George Katsikatsos in 1977. The basin of Aliveri-Kymi is a very well-studied basin (De Bruijn et al. 1980; Katsikatsos et al. 1981; Benda and De Bruijn 1982), which contains an important fossil assemblage of small mammals (De Bruijn and Van der Meulen 1979; De Bruijn et al. 1980; Van der Meulen and De Bruijn 1982; Klein Hofmeijer and De Bruijn 1985; Doukas 1986; López-Martínez 1986; Álvarez-Sierra et al. 1987; De Bruijn et al. 1987; Klein Hofmeijer and De Bruijn 1988), carnivores (Schmidt-Kittler 1983) and pollen (Benda and De Bruijn 1982).

Klein Hofmeijer and De Bruijn (1988) described the cricetids from Aliveri. They assigned the Megacricetodon material to M. primitivus after comparison with the type material from Valtorres. Nevertheless, they pointed out several differences between the material from these localities. Based on these differences in dental characteristics, they interpreted the Aliveri material as more primitive than the material from Valtorres, suggesting the earlier occurrence of Megacricetodon in the eastern Mediterranean than in the Western area. However, Oliver and Peláez-Campomanes (2014), considered that those differences between the Megacricetodon from Aliveri (specially the North quarry) and M. primitivus from Valtorres, are strong enough to place the material from these Greek localities in a new species different from M. primitivus.

Institutional abbreviations.—IAUU, Instituut voor Aard-wetenschappen, Utrecht University, The Netherlands.

Other abbreviations.—AFF, Affalterbach; AL-N, Aliveri North Quarry; AL-S, Aliveri South Quarry; ANT, Antonios; ART, Artesilla; BEZ, Bézian; COLD, La Col D; DOL3, Dolnice 3; FRZ, Franzensbad; LAN1, Langenau 1; LANG, Langenmoosen; MUN3A, Munébrega 3A; NIE, Niederaichbach; OFF2, Offingen 2; ORE, Ořechov; PEL, Pellecahus; PUTT, Puttenhausen; SE, Sérido; SS, Sandelzhausen; STR, Strakonice; THC, Thymiana; VA8A, Valdemoros 8A; VLT, Valtorres.

Material and methods

According to the locality descriptions by De Bruijn et al. (1980) Megacricetodon hellenicus sp. nov. was recovered from two different exposures of the same horizon (North and South quarries) situated in a clay bed directly underlying the lignite in the open-cast mine of Aliveri, in the Aliveri-Kymi Basin, on the Island of Evia (Greece).

The nomenclature of the cheek teeth of Megacricetodon is based on Oliver and Peláez-Campomanes (2013). The definitions of character states for morphological structures are from Oliver and Peláez-Campomanes (2013) and Oliver and Peláez-Campomanes (2014). Notation M/m is used for upper/lower molars respectively. The images of the occlusal surface of the cheek teeth were made with a Scanning Electron Microscope Fei, model Quanta 200 using a large field detector (LFD). The 3D images were made using a Micro-CT scanner Nikon XTH 160 and processed using the ImageJ software (Fiji, version 1.48r; Rasband, 1997–2014).

Measurements of the cheek teeth are after Oliver and Peláez-Campomanes (2013). They have been taken using a Nikon Kosata KK measuring microscope (10× and 15×) and are given in millimetres. Length and width represent the maximum antero-posterior and bucco-lingual distances, taken perpendicular to each other.

The specimens from Aliveri North (AL-N) and South (AL-S) quarries are stored at the IAUU.

Systematic palaeontology

Fig. 1.

Cricetodontid rodent Megacricetodon hellenicus sp. nov. from type locality Aliveri North quarry, Greece (MN4, Early Miocene). A. IAUU AL-N-2051, M1 left. B. IAUU AL-N-2052, M1 left. C. IAUU AL-N-2053, M1 left. D. IAUU AL-N-2055, M1 left. E. IAUU AL-N-2056, M1 left. F. IAUU AL-N-2057, M1 left. G. IAUU AL-N-2059, M1 left. H. IAUU AL-N-2060, M1 left. I. IAUU AL-N-2071, M2 left. J. IAUU AL-N-2072, M2 left. K. IAUU AL-N-2077, M2 right. L. IAUU AL-N-2078, M2 right. M. IAUU AL-N-2091, m1 left. N. IAUU AL-N-2092, m1 left. O. Holotype IAUU AL-N-2096, m1 right. P. IAUU AL-N-2097, m1 right. Q. IAUU AL-N-2101, m2 left. R. IAUU AL-N-2102, m2 left. S. IAUU AL-N-2104, m2 left. T. IAUU AL-N-2105, m2 left. U. IAUU AL-N-2111, m2 right. V. IAUU AL-N-2126, m3 right.

Fig. 2.

Cricetodontid rodent Megacricetodon hellenicus sp. nov. from the Aliveri South quarry, Greece (MN4, Early Miocene). A. IAUU AL - -S-1112, M1 right. B. IAUU AL-S-1114, M1 left. C. IAUU AL-S-1118, M3 left. D. IAUU AL-S-1119, M3 left. E. IAUU AL-S-1122, m1 left. F. IAUU AL-S-1124, m1 right. G. IAUU AL-S-1115, M2 left. H. IAUU AL-S-1128, m3 right. I. IAUU AL-S-1129, m3 right.

Table 1.

Measurements (in mm) of the upper and lower molars of Megacricetodon hellenicus sp. nov. from the North and South quarries, Greece (MN4, Early Miocene). TN, total number of teeth; N, number of teeth measured; min, minimum; max, maximum; σ, standard deviation.

Discussion

The Megacricetodon assemblage from the Aliveri South quarry shows similar size and morphology to M. hellenicus sp. nov. from the Aliveri North quarry. The small differences, in the relative abundance of some character states, observed between both assemblages, are interpreted as due to the scarce material available from the South quarry (see SOM 3: tables 1–25).

Despite the similar size between Megacricetodon hellenicus sp. nov. and M. primitivus, these two species show several differences: (i) M. hellenicus sp. nov. shows a dental pattern with longer and slender crests that give it a more lophodont aspect than in M. primitivus; (ii) M. primitivus has upper first molars with shorter mesoloph than M. hellenicus, and a short entoloph which posterior arm of the protocone poorly developed. By contrast, in M. hellenicus sp. nov. the mesoloph generally is of medium length and the entoloph is long and symmetrical, with tranverse sinus; (iii) M. primitivus has M2 with a higher percentage of ectolophs absent and entolophs with equal development of the posterior arm of the protocone and the anterior arm of the hypocone. In contrast, in M. hellenicus sp. nov. the ectoloph is usually long, and the entoloph is long and angular, with the posterior arm of the protocone better developed than the anterior arm of the hypocone, (iv) the upper third molars of M. primitivus show a short neo-entoloph, while in M. hellenicus sp. nov. it is well developed (based on material from Aliveri South quarry); (v) in M. primitivus the five main cusps of the m1 have similar height, whereas in M. hellenicus sp. nov. the anteroconid is notably lower than the other four cusps (see Fig. 3); (vi) the Spanish species has shorter mesolophids in the m1 and m2, while the Greek species has mesolophids normally short or medium in length.

The study by Oliver and Peláez-Campomanes (2014) of all available material of M. primitivus from the Calatayud-Montalbán Basin (type area for the species) demonstrates the remarkable morphological stability shown by this Spanish taxon along the approximately one My of its distribution in the basin. Based on the low morphological variation of M. primitivus through time, we interpret the differences between the Spanish and Greek samples (SOM 3, 4) as important enough to consider them as belonging to two different taxa, contrary to the interpretation of Klein Hofmeijer and De Bruijn (1988). These authors assigned the Megacricetodon material from Aliveri (North and South quarries) to M. primitivus, although they actually recognized and pointed out several dental characters differentiating them. Their interpretation for these morphological differences is that the samples from Aliveri are more primitive than the sample from Valtorres, suggesting the earlier occurrence of Megacricetodon in the Eastern Mediterranean than in the Western.

This earlier occurrence of Megacricetodon in Greece is supported by different authors that have studied the Greek localities and suggested an early Middle Miocene (early MN 4) for the Aliveri assemblages (Álvarez-Sierra et al. 1987; Theocharopoulos 2000; Koufos 2006). The presence of the eomyid Pseudotheridomys parvulus (Schlosser, 1884) combined with the presence of the modern cricetids such as Democricetodon, Megacricetodon, and Cricetodon has been used to assign a MN 4 age to Aliveri (Álvarez-Sierra et al. 1987). Theocharopoulos (2000), studying the Democricetodontinae, argued that the Democricetodon franconicus Fahlbusch, 1966 from Aliveri is slightly more primitive than the material from Karydia and that both are more primitive that the material from Erkertshofen 1, type locality of D. franconicus, and Petersburg 2, thus implying an earliest MN 4 for the Aliveri fauna. The relative position between these two Greek localities is further confirmed by the evolutionary stage of Cricetodon and Anomalomys, and the presence of Ligerimys instead of Pseudotheridomys in Karydia, indicating a younger position for the latter assemblage (Doukas 2003; Koufos 2006). Based on recent studies on the biostratigraphy of Spanish Aragonian and its correlation to other European basins (Van der Meulen et al. 2011, 2012), the first Megacricetodon in the Calatayud-Montalbán Basin is from late MN 4, therefore, much later than the Aliveri fauna.

Fig. 3.

Micro-CT images, showing the relative height of the five main cusps. In Megacricetodon hellenicus sp. nov. (A, IAUU AL-N-2096; B, IAUU AL-N-2097) from Aliveri South quarry, Greece (MN4, Early Miocene) the anteroconid is lower than the rest of the cusps. In Megacricetodon primitivus (Freudenthal, 1963) (C, VLT-167) from Valtorres, Spain (MN5, Middle Miocene) the anteroconid is of similar height than the other cusps. Labial (A₁–C₁) and occlusal (A₂–C₂) views.

The Aliveri association is considered to have similar age to the Czech sites of Ořechov and Dolnice 3, based on the presence of Democricetodon franconicus and Pseudotheridomys. Nevertheless, the Czech Megacricetodon cannot be assigned to M. hellenicus sp. nov. based on its larger size and different dental morphology. It is remarkable that the first representatives of Megacricetodon from other European basins belong to the Megacricetodon bavaricus group as defined by Oliver and Peláez-Campomanes (2013). The first Megacricetodon from the Swiss and German Molasse are recorded in Langenau 1 and Rauscheröd (Reichenbacher et al. 2013), and assigned to Megacricetodon aff. collongensis (Mein, 1958) and in France from Pellecahus in the Aquitanian Basin (Ginsburg and Bulot 2000), assigned to M. aff. bezianensis Bulot, 1980. All those central European forms show close similarities such as the robustness of the teeth, the slender lower first molars, the long anterolophulid, and especially the elongated anteroconid, with a “crescent” shape, of the m1, as pointed out by Oliver and Peláez-Campomanes (2013), and are clearly different in dental proportions and morphology from the Aliveri material. The morphological differences observed between the early representatives of the M. bavaricus group and M. hellenicus sp. nov. are important enough to make us think that they represent a different lineage that evolved independently.

This hypothesis is supported by the presence of similar morphological characters, in the Megacricetodon material from Aliveri and other younger Greek localities, which are not common in other Megacricetodon forms. Vasileiadou and Koufos (2005) indicated that Megacricetodon from the locality of Antonios (Macedonia, Greece) shows many similarities with Megacricetodon from Aliveri and assigned this material to M. primitivus. They also said that Megacricetodon from Antonios shows morphological similarities with M. primitivus from Valtorres. We do not agree with the allotment of Megacricetodon from Antonios neither with Megacricetodon from Aliveri, nor with M. primitivus (Oliver and Peláez-Campomanes 2014). Megacricetodon from Antonios shows a less rounded anteroconid in the m1 and the anterocone is deeply split with a small cingulum ridge in front of it, in the M1, characters that differentiate it from both M. hellenicus sp. nov. and M. primitivus. Nevertheless, it shows a particular configuration of the sinus on the upper molars that is characteristic of the Greek forms, and that differentiates it from other supposedly contemporaneous European forms like M. aff. collongensis from central Europe, M. bezianensis from France or M. primitivus from Spain.

Therefore, after the detailed study on the assemblages of Megacricetodon from Greece and other European basins (Oliver and Peláez-Campomanes 2013, 2014), important differences have been pointed out that enable us to propose the existence of at least three migration events for the early Megacricetodon forms.

Figure 4 shows the temporal and geographic distribution of early Megacricetodon taxa, indicating the proposed three main migration events. The first event corresponds to the entry of the small-sized Megacricetodon hellenicus sp. nov. recorded from Aliveri (Island of Evia) in the earliest MN 4, and that evolved as an endemic Greek form. Megacricetodon hellenicus sp. nov. dispersed and evolved through the Komotini area (Karidya I and II), Macedonia (Antonios), and Chios Island (Thymiana A and C), during the Early Miocene. A second migration event corresponds to the early representatives of the Megacricetodon bavaricus group from central European basins. The earliest representatives of these large-sized Megacricetodon forms are from Ořechov and Dolnice 3 in the Czech Republic (early MN 4) (Fejfar 1990; Abdul Aziz et al. 2010). These forms are supposedly related to the M. bavaricus group, of which the earliest representatives are Megacricetodon aff. collongensis from localities in Swiss and German Molasse, and Megacricetodon bezianensis from France (middle MN 4) (Bulot 1980). This group was slowly spreading to the west, not reaching the Iberian Peninsula until the middle of MN 5 (Oliver and Peláez-Campomanes 2013).

Fig. 4.

Biostratigraphic distribution of Early Miocene Megacricetodon species from Europe with indication of the proposed migrations events among South-eastern Europe (Greece), Central Europe (Czech Republic, Switzerland, and Germany) and South-western Europe (France and Spain), modified from Oliver and Peláez-Campomanes (2013). The silhouettes of the cricetid indicate indicate three phases of immigration of Megacricetodon species into Europe.

The third event, involves the Megacricetodon primitivus lineage. This small-sized species, only recorded from south-western Europe according to Oliver and Peláez-Campomanes (2014), appeared during MN 4, in southern France (Gers localities), after M. bezianensis (Ginsburg and Bulot 2000), and reached the Iberian Peninsula (Spain and Portugal) at latest MN 4 (Ginsburg and Bulot 2000; Oliver and Peláez-Campomanes 2014).

Conclusions

The study of the early representatives of the Megacricetodon in Europe allows us to propose a more detailed paleobiogeographic framework for the Early Miocene of Europe. The evolutionary pattern of the genus shows a configuration of three distinct Megacricetodon forms, which reach Europe in different migration waves, and with different main distribution areas. The earliest European Megacricetodon form was endemic of the eastern Mediterranean area and was small sized. It represents a new species described here as Megacricetodon hellenicus sp. nov. from the Aliveri localities. The second wave of Megacricetodon reached Central-European basins. It involves large-sized forms, included in the Megacricetodon bavaricus group, which distributed towards Western Europe till Middle Miocene. And the third form, Megacricetodon primitivus, was also small-sized and restricted so far to south-western Europe (Iberian Peninsula and southern France).