Pierre Mein Université Lyon I

On Neogene of Eurasia: distribution and migrations

Mein, P., 2003 - On Neogene rodents of Eurasia: distributions and migrations - in: Reumer, J.W.F. & Wessels, W. (eds.) - DISTRIBUTION AND MIGRATION OF TERTIARY IN EURASIA. A VOLU- ME IN HONOUR OF HANS DE BRUIJN - DEINSEA 10: 407-418 [ISSN 0923-9308] Published 1 December 2003

The most recently published papers dealing with Eurasian rodents were compiled. The importan- ce of dispersal migrations for our understanding of Eurasian faunal dynamics is stressed on the basis of selected taxa. The rodent genera appeared in Asia generally earlier than their European relatives, evidences for the way of migrations. Many migrations may not have been pre- served in the fossil record; however, most of them seem to originate in Asia, even if two or three examples show the contrary. Four main dispersal centres (China, Kazakhstan, Pakistan and Anatolia) were important during Neogene times.

Correspondence: Université Lyon I, Géode, 27-43 Boulevard du 11 Novembre, 69622 Villeurbanne Cedex France; e-mail: [email protected]

Keywords: Neogene, Rodentia, migration, Eurasia

INTRODUCTION Theridomyidae and Rhizospalacidae. Asian rodents are more diversified than Unfortunately, these two families are unk- European ones due to the smaller size of the nown in Asia, and only geomagnetic data in latter area and the smaller latitudinal varia- calibrated sections can mark the beginning of tion. Therefore, Asia acted as a faunal disper- the Neogene, except in Pakistan where the sal centre for Europe during the Neogene; the extinction of the Baluchimyinae seems con- great majority of the exchanges were migra- temporaneous with the Oligocene-Miocene tions from Asia into Europe (as in Paleogene boundary. times). Dispersal events from North America Western Europe can be seen as a trap for or Africa are rare and correlated with impor- the East-West immigrants, because of its tant tectonic or eustatic changes. Successful position at the Western end of the Eurasian immigration may result in the appearance of continent and its triangular shape. Sometimes new families on the new continent. On the the temporal and spatial distribution of some Eurasian continent however, important bar- invaders is very small. For example, the riers do not exist and small ecological varia- genus Celadensia is only known from a single tions initiate large changes in the distribution latest Miocene locality in Italy, and from of the faunas. Therefore, migrations were Teruel (Spain) during a short period at the numerous and led often to a change on genus Miocene-Pliocene transition. On the other level. hand, some immigrants settle and undergo The beginning of the Neogene in Asia is evolutionary radiation. This is illustrated by difficult to identify. In Europe, the Paleogene- the successful cricetid Democricetodon (with Neogene boundary is characterized in the its probable origin in Anatolia), which is rodents by the disappearance of the ancestral to several genera in Western

407 DISTRIBUTION AND MIGRATION OF TERTIARY MAMMALS IN EURASIA DEINSEA 10, 2003

Europe. So, Western Europe periodically Sardinian ctenodactylids. So Prosayimys received faunal invasions, but it is difficult to could be the ancestor of the Sardinian cteno- localize the sources areas, which could be dactylids. During the Miocene, ctenodactylids Eastern Europe, Central Asia, South-Western disappeared from Asia. The youngest form Asia, or the Far East. During the last decade, Sayimys chinjiensis Baskin, 1996, is known intensive field campaigns allowed to charac- from the Upper Vallesian of Pakistan. At the terize some independent dispersal centres in beginning of the Middle Miocene Sayimys Asia. Four such areas are now well known intermedius (SEN & THOMAS, 1979) migrated (corresponding to the main paleontological to Rotem (Israel), to Al Sarrar (Saudi Arabia) field activities): Mongolia, Kazakhstan, and Djebel Zelten (Lybia, Africa). Cteno- Anatolia, and Pakistan. dactyloidea were also represented during Rodents, which are the most numerous Paleogene times in Asia by the Chappa- mammals, expanded widely since the timyids. An offshoot of this last group, the Oligocene. Hartenberger (1998) gives a chart Baluchimyinae, survived until the earliest showing the major subdivisions and lineages. Miocene (Flynn et al. 1986, Flynn & Cheema During the Neogene epoch, a few new groups 1994). The recent discovery of a representati- diversified. Some of these new groups, espe- ve of Baluchimys from an Eocene level from cially the , were very successful and Thailand (Marivaux et al. 2000) has demon- produced many species and genera. McKenna strated an Eocene origin for the & Bell (1997) have realized a remarkable Baluchimyidae. review of all mammals at the supragenic Diatomyidae (Mein & Ginsburg 1997) con- level, using a strict cladistic classification. In stitute an independent Asiatic group of the this case, the Neogene innovations are limi- ctenodactyloids, they originated in Pakistan, ted to some subfamilies. This classification is and have two successive genera Fallomus not convenient, because about fifty per cent (FLYNN et al., 1986) and Diatomys LI, 1974. of the rodents are part of the huge murid Flynn (2000) shows several morphological family. In this paper, the groupings of rodents intermediate steps between these two diato- used are those that are usually employed, but myid genera. Their molars are convergent they are largely informal, because affiliations with those of Megapedetes. During the Early of many groups have not been solved satis- and Middle Miocene, this group is known factorily. I will focus on those groups whose from Laddak, Thailand, China and Japan. The migrations have been demonstrated; for exa- last occurrence is in the Pakistani site Y-797 mple, some old rodents like the aplodontids (Nagri formation; 10.6 Ma). living in China and in Central Europe during Pedetidae are typically an African rodent the early Miocene have no close relationship, family, but they are not documented during and will, therefore not be discussed. the Paleogene times. Three genera (Parapedetes, Megapedetes and a new unde- NON-MUROID RODENTS scribed genus) occurred simultaneously The Neogene European Ctenodactylidae have during the Early Miocene of East and South been recently reviewed by De Bruijn (1999), Africa. Among them, only Megapedetes can who pointed to the fact that the Sardinian cte- be occasionally found in the Middle Miocene nodactylids show a tooth structure too com- of the South Eastern Mediterranean region; plicated to be derived from a Sayimys pattern. they are known from: Rotem (Israel, MN4; But the genus Prosayimys from the Upper Wood & Goldsmith 1998), Chios (Greece, Chirtawarta formation of Zinda Pir Dome MN5; Tobien 1968) and Al Sarrar (Saudi Pakistan (estimated age 20 Ma; Baskin 1996) Arabia, MN5; Thomas et al. 1982). has preserved the metalophid II, which is Megapedetes aegaeus is the last known always present in the Early Miocene Asiatic form of the Pedetids, known from

408 MEIN: Neogene rodents of Eurasia

Bayrak Tepe 1 (MN7-8; Sen 1977). European origin is clear, because it is known Another African group, the Thryonomyidae, from Europe since MP25 and perhaps earlier. which are hystricognath rodents, occurs occa- Probably by the way of Anatolia, Dryomys sionally during the Miocene in Asia. Two dif- went to Africa during the Middle Miocene, ferent migrations reached the Indian subcon- where it can be found until the Lower tinent (Flynn & Winkler 1994). The first Turolian. The recent genus Graphiurus could with the small Kochalia DE BRUIJN, 1986, is be a descendant of this African Dryomys. known from 11.3 to 12.5 Ma and the second Myomimus was present in Pakistan from 13.8 with the large Paraulacodus HINTON, 1933, Ma (HGSP 107, Banda Daud Shah) and which is present only during a brief episode occurs now in Iran and Turkmenistan. from 12.9 to 12.5 Ma. The myrmecophagous Glirid Plioselevinia There is nothing to add to the story of the is known from MN15 of Weze (Poland) and Castoridae, described in detail by Hugueney Ivanovce (Czech Republic) and from MN16 (1999), who shows that during the Neogene, of Hajnacka (Czech Republic) (Fejfar & the same genera of this group lived in Asia Heinrich 1985). Its recent relatives live now and Europe. However, Bendukidze (1997) in Kazakhstan. So, the migration of this mentions two genera (Capacikala and group has been from West to East. Capatanka) of American origin from the Sciuroidea were widely spread upon the Early Miocene of Kazakhstan (but now, the whole Eurasiatic continent at the end of the latter is included in the genus Palaeocastor). Oligocene. During the Neogene, their diversi- Gliridae belong to an old rodent radiation fication increased. Some ground squirrels present in Eurasia since the Eocene. Many invaded Africa in the Early Miocene just after genera are exclusively European forms and the collision between African and Eurasiatic the first idea supposed that Europe was the plates. We can observe two different routes only dispersal centre for Glirids. However, for the migrations between Europe and Asia. the Oligocene-Early Miocene Glirids from The first is shown by Atlantoxerus, which is Anatolia exhibit a great number of forms and present at the beginning of Neogene of are sometimes older than the European occur- China, in the Suosuoquan Formation (MN1, rences (Ünay, 1994). For example, Glis has Qiu et al. 1999). This genus appeared in been recorded from the Middle Oligocene of Europe only in MN4 (de Bruijn 1999) and Anatolia (MP25) but is only known since persisted until the Middle Pliocene (MN15). MN4 in Europe. Similarly, Glirulus is known In Asia, the taxon disappeared after MN14 from Harami-1 (MN2) in Anatolia, and arri- (Bilike, Qiu & Storch 2000). In the same ved in Europe in MN4. Myomimus existed in group, Sciurus is supposed to arrive from Keseköy (MN3) and Çandir (MN6), but arri- North America during the Lower Pliocene. ved in MN9 in Europe (Daams 1999). During An opposite direction of migration is exem- the Early Miocene, glirids are unknown in plified by Ratufa, which is found earlier in other districts of Asia, except in China where Europe than in Asia. In Europe, this genus is a tooth of Glirulus is recorded from the Late known from Ulm Westtangente (MN2) until Oligocene of the North Junggar Basin (Wu et Obendorf (MN4, de Bruijn 1998). In Asia, al. 2000). So Anatolia may have been a good the first Ratufa is described from Li Mae centre for glirid dispersion to Europe or other Long (Thailand, MN4; Mein & Ginsburg parts of Asia. 1997). Further, this genus is mentioned from At the beginning of the Middle Miocene, several localities of the Potwar Plateau, Microdyromys, Miodyromys, Prodryomys are Pakistan, between 9.7 and 8.5 Ma (Flynn et known from Zaissan (North Asia; Kowalski al. 1995). So, the case of Ratufa is an examp- & Shevyreva 1997). Microdyromys can be le of West to East migration. found in China in MN5. For this genus, its Eomyidae have been exhaustively reviewed

409 DISTRIBUTION AND MIGRATION OF TERTIARY MAMMALS IN EURASIA DEINSEA 10, 2003

by Engesser (1999). This author shows the related genus Sinozapus is known from Bilike precise date of migration of some Neogene (MN14, Qiu & Storch 2000). So Eozapus is Eomyids: Pentabuneomys appeared in Europe earlier present in Europe than in Asia. in MN3, and Keramidomys and Eomyops Sminthozapus Sulimsky, 1962, is restricted to appeared in MN5. Unfortunately, their ance- the Pliocene of East Europe. The genus stors are unknown in Asia. This group shows Sicista first appeared during the Late a rare example of West-East migration with Miocene of Kazakhstan and migrated west- the genus Megapeomys that, original from ward to the Ukraine and eastward into China Central Europe, shows a quick expansion during the Pliocene. During the Pleistocene, reaching Japan (Fejfar et al. 1998). Sicista made some short forays related to Neogene Dipodidae of Eurasia belong to cold periods into Western Europe. two different groups: Zapodidae and Allacta- Allactaginae are first known from the latest ginae, which may represent two different Oligocene of Kazakhstan and Anatolia. families. McKenna & Bell (1997) recognize Among the Miocene representatives, the only one family, but divide the Sicistinae genus Protalactaga had a wide Asian distri- from the Zapodinae at subfamily rank. For bution during Early and Middle Miocene. It European researchers, the Sicistinae and reached North Africa at the end of the Middle Zapodinae are more related to each other than Miocene, but its African range is very restric- either is to the Allactaginae. The dipodids ted due to its extinction just before the arrival have recently been revised by Daxner-Höck of Hipparion. Protalactaga gabunii (LUNGU, 1999. 1981) is an European allactagine, found in In Europe, Plesiosminthus, a migrant des- the Vallesian of Moldavia; it may be ancestral cendant from the Oligocene Chinese to Lophocricetus, an important genus during Parasminthus, is known from the Late the Turolian of the Ukraine, Siberia and Oligocene (MP26) to the Early Miocene China. Paralactaga from the latest Miocene (MN2). A second immigration of an unknown of China and Kazakhstan seems to be the zapodid occurs in MN3/4 (Petersbuch, ancestor of Allactaga, which had a wide dis- Central Europe; Daxner-Höck 1999). The tribution in Asia since the Pliocene and made third immigration was that of Eozapus in the some incursions in the Ukraine and Romania Late Miocene. This tiny is known during the Late Pliocene and the Early from about twenty localities of Europe Pleistocene. between MN10 and MN14. In Spain, its tem- poral distribution is divided into two: the first MUROIDS at the boundary MN10-MN11 within locali- The Paleogene Paracricetodontinae has one ties as Cucalon and Cascante-Cubla, the surviving genus during the Early Miocene. second one in the earliest Pliocene (Celadas This genus, Mirabella DE BRUIJN et al., 1987, 3-4-5-8). The latter was incorrectly assigned appears in Anatolia (M. anatolica DE BRUIJN to Sminthozapus (Mein et al. 1983) as poin- & SARAÇ, 1992). A second species occurs in ted out by Fahlbusch (1992). In France, Keseköy at MN3. Later, the genus migrated Eozapus is recorded in the MN10 faunas of into Greece in MN4 where it is known from Amberieu 1 and 2, and Dionay. No occurren- Aliveri and Karydia. It is not recorded from ces are known from MN11 and MN12, but younger faunas, so its temporal range is rela- this genus reappeared in Lissieu MN13 (Mein tively short. De Bruijn & Saraç (1992) note 1999) and also in Celleneuve MN14. In the remarkable fact that there were probably Eastern Europe, Eozapus was found in two different migrations during the short Berislave (Nesin & Topachevsky 1999) and range of the genus into Central Europe each Podlesice (MN14). In China, Eozapus is not one evidenced by only a single molar: a large present before MN13 (Fahlbusch 1992) and a one from the MN3 fauna of Reiden-Sertel

410 MEIN: Neogene rodents of Eurasia

(Swizerland) and a smaller one from the Kentjubeck (MN6), can be easily assigned to MN4 fauna of Rembach (Germany). This Cricetodon, but the type species T. irtyschen- example illustrates that many migrations may sis Topachewski & Skorik, 1988 from the not have been preserved in the fossil record. Vallesian locality of Semipalatynsk region, is The Cricetodontinae are an important group a more derived form: brachyodont, but with a of Muroids because of their wide geographic flat occlusal surface and cement filling the and long temporal distribution. Our knowled- synclines; it is one of the numerous forms to ge about them has significantly increased, be convergent with voles. (The Early particularly through the fieldwork of Hans de Miocene age assignment of this genus in Bruijn’s team. The nominative genus Criceto- McKenna & Bell (1997) is erroneous). don LARTET, 1851, has first been recorded Zramys JAEGER et al., 1973, spread over from the Middle Miocene of Western Europe. North Africa during the final Middle Miocene De Bruijn et al. (1993), have shown its exis- and Late Miocene. Probably this migration tence in the Aegean region and Anatolia. The came from Southwestern Asia, via Anatolia. first representatives of the genus are characte- At the end of the Miocene, another migration rized by three roots on first and second upper (with a Spanish origin) introduced molars and a small size. The oldest known Ruscinomys Depéret, 1890, into North Africa Cricetodon has been found at Inkonak across the Gibraltar strait. In Western and (MP30, de Bruijn et al. 1991); from MN1 to Central Europe, other Middle Miocene crice- MN3 in Anatolia Cricetodon was present, so todontines have been found besides this region should have been the dispersal Cricetodon. The largest one is Lartetomys centre for all Cricetodontids. Greece has been MEIN & FREUDENTHAL, 1971, which has three colonized during MN4 and Central and roots in its upper molars. L. mirabilis is a Western Europe during MN5. During the very rare species occurring in France during Middle Miocene, the genus Cricetodon is MN5. Mixocricetodon dehmi RUMMEL, 1997 known from several Turkish localities like (well represented in several German fissure Pasalar, Çandir, Sariçay; it is also known fillings), is a junior synonym of Lartetomys. from the MN6 fauna of Belometschetskaya Recently, one isolated upper second molar (North Caucasia, the type locality of C. cau- has been recovered from the new French sasicus PICKFORD et al., 2000). Cricetodon locality Four (MN6/7, Maridet et al. 2000). appeared early in China: Cricetodon sp. is This tooth (LxW=3.31x2.68 mm) is more mentioned from the MN1 locality of recent and more voluminous than the upper Suosuoquan (Qiu et al. 1999) and from second molar of L. mirabilis. L. mirabilis and Sihong during MN4 (Qiu 1996). T. irtyschensis are the largest known criceto- Gobicricetodon QIU, 1996 appeared during dontines. Is a lineage between these two spe- the late Middle Miocene and Early Vallesian; cies possible through the species of Four? A this form has a size similar as the Western- second species, described as Lartetomys European species, but still has three roots on zapfei, has been now correctly interpreted as its upper molars. Another Chinese genus a new genus Karydomys of the Democri- Plesiodipus YOUNG, 1927 from the Middle cetodon-Copemys group (Theocharopoulos, Miocene is probably related to Gobi- 2000). cricetodon, and at the base of the Following the opinion expressed by de Myospalacine radiation during the Late Bruijn et al. (1993), the genera Deperetomys Miocene. and Eumyarion can be included in the In Kazakhstan, the genus Tsaganocricetus Eucricetodontinae. Besides the morphological TOPACHEWSKI & SKORIK, 1988, is represented arguments, these two genera share also the by two different species. The older one, T. same dispersal centre Anatolia and have the turgaiensis BENDUKIDZE, 1993 from same first appearance date. Deperetomys

411

DISTRIBUTION AND MIGRATION OF TERTIARY MAMMALS IN EURASIA DEINSEA 10, 2003

MEIN & FREUDENTHAL, 1971, has been firstly lities of Anatolia. recorded from the latest Middle Miocene of Another muroid from Anatolia is Muhsinia. Central and Western Europe. In Anatolia, new This new genus was found at Inkonak (MP30) and older species have been described by de and Kilçak 0 (MN1, de Bruijn et al. 1992). It Bruijn et al. (1993) from Kargi II (MP30) was tentatively positioned inside the group of and Kilçak 0 (MN1), and the younger MN4 the Pseudocricetodontidae by the authors, and locality of Gemerek. This latter find links the finally stored into the Eucricetodontini by first species from Anatolia to the most recent McKenna & Bell (1997). This genus bears species from Europe. Meteamys de Bruijn et some likeness to Afrocricetodon LAVOCAT, al., 1992 is a large Cricetodontinae closely 1977. This strange genus may be at the origin resembling Deperetomys, but its size in the of an Afrocricetodontine migration into latest Oligocene of Inkonak and Kargi is as Africa. It is a very low crowned muroid, with large as the Deperetomys of the late Middle longitudinal ribs on the lower incisor and two Miocene of Europe. metalophids on lower molars. Eumyarion THALER, 1966 is a genus known Democricetodontinae (or Copemyinae) are in Europe from MN4 to MN9. In Anatolia, one of the most frequent Neogene muroid older species were described by de Bruijn & rodents. The main genus Democricetodon Saraç (1991) from MN1 of Harami to MN3 FALBUSCH, 1964 is known as an Eastern of Keseköy. Additionally, Eumyarion is al- immigrant in Europe, where its arrival indica- ready present in MP30 from Kargi (de Bruijn tes the beginning of MN4. As indicated by in Theocharopoulos 2000). As in the case of the etymology of its name, Democricetodon Deperetomys, Anatolia probably is the first is a very common animal, often the dominant dispersal centre of Eumyarion. The species genus in small- faunas. Recently, found in this district are good ancestors for Theocharopoulos (2000) has reviewed the the European burrowing rodents, the Anoma- complete history of this group. Like so many lomyinae. Eumyarion appears also in Pakistan muroid rodents, the origin of this group with E. kowalskii (Lindsay 1996), in the seems to have taken place in Anatolia descen- Chirtawata Formation of Zinda Pir Dome, ding from the Middle Oligocene genus calibrated to 20 Ma (early MN3). This last Lignitella ÜNAY. Democricetodon itself appe- species seems a good ancestor for the rhizo- ars in the earliest Neogene (Kilçak 0) and is myid radiation. Wessels & de Bruijn (2001) present in all known Early Miocene localities transfer this species to the genus Proka- and some Middle Miocene localities until nisamys. Later, the genus Aralocricetodon MN6 from Anatolia. Çandir is the last known BENDUKIDZE, 1993 shows an ephemerous Turkish site where this genus present. invasion of Rhizomyidae during the Middle In Europe, Democricetodon became less Miocene of Kazakhstan. So two different frequent in the Late Miocene in relation with kinds of burrowing rodents originated from the presence of new (but related) genera. The Eumyarion. last occurrences are in the early MN10 locali- A third group of specialized burrowing ties such as Masia del Barbo (Spain) and rodents are the Spalacidae. This family living Suchomasty (Czech Republic). In Pakistan, in South West Asia and South East Europe Democricetodon is known from numerous has been recently well studied by Ünay species (Lindsay & Downs 1998) from Y-747 (1999). The fossil record of this family seems (18 Ma) to Nagri levels or perhaps later. In to indicate Anatolia as the centre of origin China (Qiu et al. 1999), the first occurrence and dispersion. The oldest known spalacid is is in the Sihong fauna, which is chronologi- Debruijnia from Keseköy (Anatolia, MN3), cally at the same level as in Europe, posterior but until now, no potential ancestor has been to the fauna of Tunggur (MN7/8). The last found among the faunas from the older loca- occurrence is from Amuwusu (Nei Mongol,

412 MEIN: Neogene rodents of Eurasia

early MN9). ce of the same genus is known from Bayrak- In Anatolia, Democricetodon was preceded tepe 1 (MN7/8, Anatolia). Cricetines genera by the genus Spanocricetodon LI, 1977. S. were very numerous during the Late Pliocene sinuosus THEOCHAROPOULOS, 2000, is known and Pleistocene. The main origin of these from the latest Oligocene of Inkonak and lies certainly in the vast open areas from some MN1 localities. Later, this genus of Central Asia. migrated to Pakistan where S. khani DE Two different kinds of specialized ‘cricet- BRUIJN et al., 1981, has been found in HGSP ids’ occurred in the Late Miocene. The first 116 from the Murree Formation (MN3), and are the so-called ‘gerbilloids cricetids’ with in the coeval Zinda Pir Dome Y-113 (Lindsay three genera showing migrations: Blancomys 1998). Subsequently, Spanocricetodon expan- WEERD et al., 1977 has been found in Spain ded throughout Asia during MN4, as shown and France between MN13 and MN16, but it by the occurrence of S. ningensis (the type was already present in Anatolia in the MN10 species) in Fangshan (China) and S. janvieri locality of Karaözü and Kaleköy (Sümengen MEIN & GINSBURG, 1997, in Thailand. et al. 1989). Epimeriones DAXNER-HÖCK, Spanocricetodon became extinct during 1972, is known only from Europe, but its MN4, it is considered to be an ancestral form geographical range is expanding eastward to Democricetodon, Primus and perhaps also during the Late Neogene. It is known from Megacricetodon. Theocharopoulos is the first MN11 to MN13 in Central and Western paleontologist to write that Protarsomys Europe, and its more recent occurrences are LAVOCAT, 1977, was a member of the in Eastern Europe till the lowermost Democricetodon group that migrated into Pleistocene (Terzea 1978). Africa. This genus is known from the Early Pseudomeriones SCHAUB, 1934 was consi- Miocene of the East Africa and Namibia. A dered as an aberrant gerbillid, but today, it is second migration wave (from Anatolia?), regarded as a gerbilloid cricetid like invaded East Africa in the late Middle Epimeriones. The oldest records are from Miocene. several Vallesian localities in Anatolia Karydomys is a new genus erected by (Sümengen et al. 1989). Later a Turolian Theocharopoulos (2000) for a special demo- migration brought this form to South East of cricetodontine lineage, characterized by a low Europe, where it is found until the Early anteroconid on m1, which is blade like and Pliocene. During the Turolian, it is known very close to the main anterior cusps. Only also from Afghanistan and China, where it one species is known from Europe described stayed until the Early Pliocene (Qiu & Storch by Mein & Freudenthal as a second 2000). The last record is found from Çalta Lartetomys species: L. zapfei, occuring in (MN15) in Western Asia; but the most intri- MN5 and MN6. Now, two different and smal- guing occurrence is from the Spanish locality ler species have been described from the of Casablanca-M, the only record in Western MN4 of Greece: K. symeonidisi and K. bos- Europe of final MN13 age; it probably arri- kosi. An Asiatic origin is likely because some ved there by way of North Africa (Agustí species were present in MN4 of Anatolia and 1989). also in MN5 of Kazakhstan. Another group of democricetodontine Around the Middle to Late Miocene transi- rodents can be called “microtoid cricetids” tion, some new democricetodontines appe- evolving toward voles. They comprise nume- ared which are considered as the beginning of rous and independent lineages. Fejfar (1999) the Cricetinae. Cricetulodon HARTENBERGER, gives a good overview of these forms, distin- 1965, is the oldest representative. This spe- guishing several morphological categories. cies is a good marker for the beginning of Microtocricetus FAHLBUSCH & MAYR, 1975, MN9 in Western Europe. An earlier occurren- is the oldest European form whose molars

413

DISTRIBUTION AND MIGRATION OF TERTIARY MAMMALS IN EURASIA DEINSEA 10, 2003

show a tendency toward prismatic structure. their synclines; the enamel free areas on the This genus lived only in MN9, from the lateral walls of the teeth increased, and the Ukraine to France. Rotundomys MEIN, 1965, thickness of the occlusal enamel became dif- is less specialized and known essentially ferentiated. The number of known forms from MN10 in Western Europe only. Microto- increased and the story continued till the scoptes SCHAUB, 1934, has hypsodont molars Pleistocene when forms with rootless molars with a prismatic structure. It is present in appeared. This group, showing rapid morpho- MN12 at Cherevichnoe (Ukraine), from logical evolution of their teeth, is the most MN13 of Central Asia and MN14 of China. useful for detailed biochronology. This genus is also known from North Megacricetodontinae seem to appear simul- America during the Lower Hemphilian taneously (MN3) in two Asian regions: in (which seems to be an older age). Pakistan (Zinda Pir Dome; Lindsay & Downs Ischymomys ZAZHIGIN, 1992, which shows a 1998) and in Anatolia, (Keseköy, Theocharo- similar dental structure, is known from the poulos 2000). In the latter locality, two diffe- MN11 fauna from Frunzkovaz (Ukraine) and rent species co-occur (de Bruijn, in Theocharo- from MN12 of Petropavlosk (Kazakhstan). poulos). Megacricetodon FAHLBUSCH, 1964, One last group with mesodont molars and a arrived in China in the Sihong fauna (MN4) trilophodont m1 consists of numerous genera, and reached Europe at the same time when probably all derived from Microtodon in this genus constitutes a major element among Eastern Asia from MN13 and MN14; some rodent faunas till MN9. In Europe, Megacrice- genera (like Trilophomys DEPÉRET, 1892 and todon disappeared without any descendants; Baranomys KORMOS, 1933) show a relative however in Pakistan, it gave rise to a larger long temporal range during the Pliocene of genus Punjabemys LINDSAY, 1988 and proba- Europe (from MN14 to MN16). Some others bly, some relative form was ancestral to the have had an extremely short temporal range, Myocricetodontinae (Gerbillidae) whose first such as Celadensia MEIN et al., 1983 that representatives are difficult to distinguish lived in Spain at the MN13-MN14 transition. from Megacricetodon. On this subject, see Bjornkurtenia KOWALSKI, 1992 and Baranar- the divergent opinions of Wessels (1996) and viamys NESIN, 1996 have similar short ran- LINDSAY & DOWNS (1998). A large review of ges. gerbils can be found in Wessels (1998, 1999). Aratomys ZAZHIGIN (in Gromov & Poljakov Myocricetodontinae arose probably in 1977) is known from the Eastern Asian loca- Pakistan; during the Early Miocene, they lities Tchurno-kariakh (Mongolia, MN13) and invaded Africa via Arabia. Known from Bilike (inner Mongolia, MN14). At first con- North Africa, but also from Kenya and sidered as a true vole by Fejfar (1997), this Namibia during the Middle Miocene, some form has been placed in the cricetids by Qiu forms reappeared in Spain during the salinity & Storch (2000). crisis (latest MN13). Calomyscus THOMAS, Arvicolidae, or true voles, make up the 1905 is a recent survivor of the Myocriceto- most recent radiation among Muroid rodents; dontinae; some fossil specimens have been after their origin in North America, they found in the uppermost Miocene of France rapidly spread over Asia and Europe during and Spain, as well as in the lower Pliocene of the Pliocene. Promimomys KRETZOI, 1955 is Rhodes. the only genus present in Eurasia during Taterillinae (Gerbillidae) arose with two MN14, with two successive species (Fejfar et close genera: the first, Abudhabia DE BRUIJN al. 1997). In MN15, Mimomys F. MAJOR, & WYBROW, 1994, originated in the Early 1902 and Dolomys NEHRING, 1898 appeared Turolian of Dhok Pathan formation (Flynn & in Western Europe at about 4 Ma. From Jacobs 1999) and is later known from Saudi MN16 onwards, the teeth have cement in Arabia and Afghanistan. The second fossil

414 MEIN: Neogene rodents of Eurasia

genus is Protatera JAEGER, 1977; it is a North had a large distribution, being known from African genus known from the Middle and East Asia to Western Europe: it is the case for Late Miocene. Some gerbils referred to the Rhagapodemus, Parapodemus and Micromys; genus Debruijnimys CASTILLO & AGUSTÍ, the latter is still living today. Finally, some 1996, are known from the lower Pliocene of genera are known from Asia and from Africa Spain and had a North African origin. Murids like Saidomys and Mus. Murids spread quick- also arose in Pakistan. We can imagine a suc- ly on Africa during the Late Miocene, espe- cession of morphological transformations cially for the group of Acomys. from Megacricetodon to Myocricetodontines, and finally to Murids. During these morpho- CONCLUSIONS logical changes of the molars, the longitudin- During the Neogene the main migrations of al crest disappeared, the alternate major cusps rodents were from Asia to Europe, especially took a twin position and new cusps appeared from Anatolia; this region has been a very on the lingual border of the upper and labial important centre for speciation before MN4 side of the lower teeth. (Cricetodon, Democricetodon, Karydomys, Dendromurines are essentially an African for example) and an important centre of dis- group with some remnants of the longitudinal persion to Europe by the South Aegean crest and only one supplementary lingual region. The same applies to Africa: dispersals cusp (t4 = enterostyle). The only Asian rodent from Anatolia (across Arabia) were more considered by Lindsay as a dendromurid is numerous and more various than from other Potwarmus LINDSAY, 1988. This genus is regions. We mention for the Early Miocene: known from Pakistan and Thailand. Its tem- Protarsomys, Muhsinia, and some ground poral range is from 18-14.3 Ma (Flynn et al. squirrels; for the early Middle Miocene: 1995). This genus migrated to Africa and was Sayimys; for the late Middle Miocene: found in Gebel Zelten (Lybia) at the begin- Dryomys, Democricetodon, Myocricetodon; ning of the Middle Miocene. and for the Late Miocene the murids Mus, Antemus JACOBS, 1978, is the first murid Saidomys and Acomys. A lower number spread showing the new cusp (t1 = anterostyle), from East Asia (eventually, from North which characterizes the Murinae. This form is America), or Central Asia. Migrations from known from 13.7-12.5 Ma (Flynn et al. Africa are only related to the Mediterranean 1995). Progonomys SCHAUB, 1938, is the first crisis, at the end of the Miocene through the murid in Western Europe appearing at the Gibraltar strait. beginning of MN10 (approximately 9.7 Ma). West-East migrations from Europe to Asia It appears earlier in Pakistan where finds of are seldom, but some have been assumed: we teeth are mentioned at 12.1 Ma (Flynn et al. mentioned Ratufa, Microdyromys, Pliose- 1995). Progonomys hussaini CHEEMA et al., levinia, Megapeomys and Eozapus. Beside 2000 is also a primitive murid present these so important exchanges between Asian between 10 and 11 Ma at the Potwar plateau. and European faunas, some groups like the Some recently known archaic murids from Tachyoryctoidinae or Myospalacinae are Sinap Tepe I (MN9) are being studied by exclusively Asiatic; they are limited to one Sevket Sen. These finds implicate an earlier continent and underwent no expansion phase. arrival of murids in Western Asia than in With the biochronological development Europe. During the Late Miocene and parti- among mammals, the first appearance follo- cularly during the Pliocene, the number of wing a migration becomes more and more murid genera increased considerably. Some important. These first appearance data are fossil genera are known only from Asia, like really excellent in many cases but when the Ratchaburimys, Chardinomys, Wushanomys, new districts are far from the original centre, Dianomys, Orientalomys. Some others have some lateness may occur during the dispersal

415

DISTRIBUTION AND MIGRATION OF TERTIARY MAMMALS IN EURASIA DEINSEA 10, 2003

event and a magnetostratigraphical control is Rössner, G.E. & Heissig, K. (eds.) - The Miocene desirable. Land of Mammals of Europe – pp. 271-280, Verlag Dr F. Pfeil, München ACKNOWLEDGEMENTS De Bruijn, H., Fahlbusch, V., Saraç, G. & Ünay, E., 1993 I would like to thank the two reviewers G. - Early Miocene rodent faunas from the eastern Storch and A. van der Meulen for their judi- Mediterranean area. Part III. The genera Deperetomys cious remarks on the scientific content and and Cricetodon with a discussion of the evolutionary especially the latter who has tried to make history of the Cricetodontini - Proceedings of the my English version more suitable. Koninklijke Nederlandse Akademie van Wetenschappen 96 (2): 151-216 REFERENCES De Bruijn, H. & Saraç, G., 1992 - Early Miocene rodent Agustí, J., 1989 - On the peculiar distribution of some fauna from the Eastern Mediterranean area. Part II. muroid taxa in the Western Mediterranean region - in: Mirabella (Paracricetodontinae, Muroidea) - Torre, D. (ed.) - Continental faunas at the Miocene Proceedings of the Koninklijke Nederlandse /Pliocene boundary - pp. 147-154, Bollettino de la Akademie van Wetenschappen 95(1): 25-40 Societa Paleontologica Italiana De Bruijn, H., Ünay, E., Hoek Ostende, L.van der & Baskin, J.A., 1996 - Systematic revision of Ctenodac- Saraç, G., 1992 - A new association of small mam- tylidae (Mammalia, Rodentia) from the Miocene of mals from the lowermost Lower Miocene of Central Pakistan - Paleovertebrata 25 (1): 1-49 Anatolia - Geobios 25(5): 651-670 Bendukidze, O.G., 1997 - The Oligocene Rodents of De Bruijn, H., Ünay, E. & Hoek Ostende, L.van der, Central and Western Kazakhstan and their stratigrap 1996 - The composition and diversity of small mam- hic significance – in: Aguilar, L.S. & Michaux, J.-P. mal associations from Anatolia through the Miocene (eds.) - Actes du Congrès BiochroM'97, Biochro- – in: Bernor, R.L., Fahlbusch, V. & Mittmann, H.-W. nologie mammalienne du Cénozoïque en Europe et (eds.) – The Evolution of Western Eurasian Neogene domaines reliés – pp. 205-208, Ecole Pratique des Mammal Faunas, pp. 266-270, Columbia University Hautes Etudes Sciences de la Vie et de la Terre, Press, New York Montpellier De Bruijn, H. & Von Koenigswald, W., 1994 - Early Cheema, I.U., Raza, S.M., Flynn, L.J., Rajpar, A.R. & Miocene rodent faunas from the Eastern Mediterra- Tomida, Y., 2000 - Miocene small mammals from nean area. Part V. The genus Enginia (Muroidea) Jalapur, Pakistan and their biochronological implica- with a discussion of the incisor enamel - Proceedings tions- Bulletin of the National Science Museum of the Koninklijke Nederlandse Akademie van Tokyo C 26 (1-2): 55-77 Wetenschappen 97(4): 381-405 Daams, R., 1999 - Family Gliridae – in: Rössner, G.E. & Engesser, B., 1999 - Family Eomyidae – in: Rössner, Heissig, K. (eds.) - The Miocene Land of Mammals G.E. & Heissig, K. (eds.) - The Miocene Land of of Europe – pp. 301-318, Verlag Dr F. Pfeil, München Mammals of Europe – pp. 319-335, Verlag Dr F. Daxner-Höck, G., 1999 - Family Zapodidae – in: Pfeil, München Rössner, G.E. & Heissig, K. (eds.) - The Miocene Fahlbusch, V., 1992 - The Neogene mammalian faunas of Land of Mammals of Europe – pp. 263-266, Verlag Ertemte and Harr Obo in Inner Mongolia (Nei Dr F. Pfeil, München Mongol), China - Senckenbergiana Lethaea 72: De Bruijn, H., 1998 - Vertebrates from the Early 199-217 Miocene lignite deposits of the opencast mine Fejfar, O., Heinrich, W.D., Pevzner, M.A. & Vangen- Oberdorf (Western Styrian basin, Austria) - Annalen geim, E.A., 1997 - Late Cenozoïc sequences of mam- des Naturhistorischen Museums Wien 99: 99-137 malian sites in Eurasia: an updated correlation – De Bruijn, H., 1999a - Superfamily Ctenodactyloidea – Palaeogeography, Palaeoclimatology, Palaeoecology in: Rössner, G.E. & Heissig, K. (eds.) - The Miocene 133: 259-288 Land of Mammals of Europe – pp. 263-266, Verlag Fejfar, O., Rummel, M. & Tomida, Y., 1998 - New Dr F. Pfeil, München Eomyid genus and species from the Early Miocene De Bruijn, H., 1999b - Superfamily Sciuroïdea – in: (MN zones 3-4) of Europe and Japan related to

416 MEIN: Neogene rodents of Eurasia

Apeomys (Eomyidae, Rodentia, Mammalia) - the Miocene: phylogenetic consequences - Comptes National Science Museum Monographs Tokyo 14: Rendus de l'Académie des Sciences de Paris 326: 123-143 439-444 Fejfar, O., 1999 - Microtoid Cricetids - in: Rössner, G.E. Hugueney, M., 1999 - Family Castoridae - in: Rössner, & Heissig, K. (eds.) - The Miocene Land of Mammals G.E. & Heissig, K. (eds.) - The Miocene Land of of Europe – pp. 365-372, Verlag Dr F. Pfeil, München Mammals of Europe – pp. 281-300, Verlag Dr F. Fejfar, O. & Heinrich, W.-D., 1985 - Importance of two Pfeil, München sites of fossil vertebrates, Ivanovce and Hajnacka, for Kowalski, K., 1979 - Fossil Zapodidae (Rodentia, mammalian paleontology in European Pliocene and Mammalia) from the Pliocene and Quaternary of Early Pleistocene: present stage of knowledge and Poland - Acta Zoologica Cracoviensia 23 (9): 697- problems - Vestnik Utredniho ustavu geologického 703 60(4): 213-224 Kowalski, K. & Shevyreva, N.S., 1997 - Gliridae Flynn, L.J. & Cheema, I.U., 1994 - Baluchimyine (Mammalia: Rodentia) from the Miocene of the rodents from the Zinda Pir Dome, W. Pakistan: syste- Zaisan Depression (Eastern Kazakhstan) - Acta matic and Biochronologic implications - in: Tomida, Zoologica Cracoviensia 40 (2): 199-208 Y., Flynn, L.J. & Jacobs, L.L. (eds.) - Rodent and Lindsay, E.H. & Downs, W.R., 1998 - Cricetid Rodents Lagomorph families of Asian origins and diversifica- from Miocene deposits of Pakistan - Geological tion – pp. 115-129, National science Museum survey of Pakistan, Islamabad: 35-47 Monographs, Tokyo Maridet, O., Berthet, D. & Mein, P., 2000 - Un nouveau Flynn, L.J. & Winkler, A.J., 1994 - Dispersalist implica- gisement karstique polyphasé miocène moyen "Four" tions of Paraulacodus indicus: a South Asian rodent (Isère, France): étude des (Mammalia, of African affinity - Historical Biology 9: 223-235 Rodentia) - Géologie de la France 2: 71-79 Flynn, L.J., Barry, J.C., Morgan, M.E., Pilbeam, D., Marivaux, L., Bennami, M., Ducrocq, S., Jaeger, J.J. & Jacobs, L.L. & Lindsay, E.H., 1995 - Neogene Chaimanee, Y., 2000 - A new baluchimyine rodent Siwalik mammalian lineages: Species longevities, from the Late Eocene of the Krabi Basin (Thailand): rates of change and modes of speciation - Palaeo- palaeobiogeographic and biochronologic implications geography Palaeoclimatology Palaeoecology 115: - Comptes Rendus de l'Académie des Sciences de 249-264 Paris, Earth and Planetary Sciences 331: 427-433 Flynn, L.J., Downs W., Morgan, M.E., Barry, J.C. & McKenna, M. & Bell, S.K., 1997 - Classification of Pilbeam, D., 1998 - High Miocene species richness in mammals above the species level - Columbia the Siwaliks of Pakistan - in: Tomida, Y., Flynn, L.J. University Press, New York, 631 pp. & Jacobs, L.L. (eds.) - Rodent and Lagomorph fami- Mein, P., Moissenet, E. & Adrover, R., 1984 – lies of Asian origins and diversification – pp. 167- L’extension et l’âge des formations continentales 180, National science Museum Monographs, Tokyo pliocènes du fossé de Teruel (Espagne) - Comptes Flynn, L.J., Downs W., Opdyke, N., Huang, K., Lindsay, Rendus de l’Académie des Sciences de Paris Série II E., Ye, J. & Xie, G., 1999a - Recent advances in the 339: 1603-1610 small mammal Biostratigraphy and Magnetostraphy Mein, P., 1999 - The Late Miocene small mammal suc- of Lanzhou Basin - Chinese Science Bulletin 44 cession from France, with emphasis on the Rhône (suppl.): 105-117 Valley localities – in: Agustí, J., Rook, L. & Andrews, Flynn, L.J. & Jacobs, L.L., 1999b - Late Miocene small- P. (eds.) - Hominid Evolution and Climatic change in mammal faunal dynamics: the crossroads of the Europe – pp. 133-157, Cambridge University Press, Arabian Peninsula – in: Whybrow, P.J. & Hill, A. Cambridge (eds.) - Fossil vertebrates of Arabia – pp. 412-419, Mein, P. & Ginsburg, L., 1997 - Les Mammifères du Yale University gisement miocène inférieur de Li Mae Long Flynn, L.J., 2000 - The great small mammal Revolution - (Thaïlande): systématique, biostratigraphie et Himalayan Geology 21 (1): 1-4 paléoenvironnement – Geodiversitas 19(4): 783-843 Hartenberger, J.L., 1998 - Description of the radiation of Nesin, V.A., 1996 - An ancient fossil vole species the Rodentia (Mammalia) from the Late Paleocene to (Rodentia, Cricetidae) from the lower Pontian South

417

DISTRIBUTION AND MIGRATION OF TERTIARY MAMMALS IN EURASIA DEINSEA 10, 2003

Ukraine - Vestnik zoologii Kiev 3: 74-75 Jahrbuch der Vereinigung ‘Freunde der Universität Nesin, V.A., & Topachevsky, V.A., 1999 - The Late Mainz’: 51-58 Miocene small mammal succession in Ukraine - in: Topachevsky, V.A. & Skorik, A.F., 1984 - The first find Agustí, J., Rook, L. & Andrews, P. (eds.) - Hominid of a Lophiomyinae (Rodentia, Cricetidae) fossil Evolution and Climatic change in Europe – pp. 265- representative - Vestnik Zoologii (Zoological Record) 272, Cambridge University Press, Cambridge 2: 52-57 Pickford, M., Gabunia, L., Mein, P., Morales, J. & Topachevsky, V.A. & Skorik, A.F., 1988 - New Vole- Azanza, B., 2000 - Belometchetskaya, North toothed Cricetodontidae (Rodentia, Cricetidae) from Caucasus, an important biostratigraphic link between Vallesian of Eurasia and some questions of super- Europe and China - Geobios 33 (2): 257-267 generic systematic of the family - Vestnik zoologii / Qiu, Z.D., 1996 - Middle Miocene micromammalian Organ Instituta zoologii Akademii nauk Ukrainskoj fauna from Tunggur - Beijing Science 1: 1-216 SSR. Kiev 5: 33-45 Qiu, Z.D. & Storch, G., 2000 - The Early Pliocene Ünay, E., 1994 - Early Miocene rodent faunas from the Micromammalian Fauna of Bilike, Inner Mongolia, eastern Mediterranean area. Part IV. The Gliridae - China (Mammalia: Lipotyphla, Rodentia, Lago- Proceedings of the Koninklijke Nederlandse morpha) - Senckenbergiana lethaea 80 (1): 173-229 Akademie van Wetenschappen B 97(4): 445-490 Qiu, Z.X., Wu, W.Y. & Qiu, Z.D, 1999 - Miocene Ünay, E., 1999 - Family Spalacidae - in: Rössner, G.E. & Mammal faunal sequence of China: Palaeozoo- Heissig, K. (eds.) - The Miocene Land of Mammals geography and Eurasian relationships - in: Rössner, of Europe – pp. 421-425, Verlag Dr F. Pfeil, München G.E. & Heissig, K. (eds.) - The Miocene Land of Wahlert, J.H., 1984 - Relationships of the Extinct Rodent Mammals of Europe – pp. 443-455, Verlag Dr F. Cricetops to Lophiomys and the Cricetinae (Rod. Pfeil, München Cricetidae) - American Museum Novitates 2784: 1-15 Sen, S., 1977 - Megapedetes aegaeus n.sp. (Pedetidae) et Wessels, W., 1996 - Myocricetodontinae from the à propos d’autres ‘rongeurs africains’ dans le Miocene of Pakistan - Proceedings of the Koninklijke Miocène d’Anatolie - Geobios 10 (6): 983-986 Nederlandse Akademie van Wetenschappen 99 (3-4): Sümengen, M.E., Ünay, E., Saraç, G., Bruijn, H. de, 253-312 Terlemez, I. & Gürbüz, M., 1990 - New Neogene Wessels, W., 1998 - Gerbillidae from the Miocene and rodent assemblages from Anatolia (Turkey) – in: Pliocene of Europe - Mitteilungen der Bayerischen Lindsay, E.H, Fahlbusch, V. & Mein, P. (eds.) - Staatssamlung für Paläontologie und Historische European Neogene Mammal chronology – pp. 61-72, Geologie München 38: 187-207 NATO ASI Series A Wessels, W., 1999 - Family Gerbillidae - in: Rössner, Terzea, E., 1978 - Epimeriones dacicus n. sp. (Rodent, G.E. & Heissig, K. (eds.) - The Miocene Land of Mammalia) du Villafranchien supérieur de Roumanie Mammals of Europe – pp. 395-400, Verlag Dr F. - Travaux de l’Institut de Spéologie ‘E. Racovitza’ Pfeil, München Bucaresta 17: 135-138 Wessels, W. & Bruijn, H. de, 2001 - Rhizomyidae from Theocharopoulos, C., 2000 - Late Oligocene-Middle the lower Manchar Formation (Miocene, Pakistan) - Miocene Democricetodon, Spanocricetodon and Annals of Carnegie Museum 70 (2): 143-168 Karydomys n. gen. from the Eastern Mediterranean Wood, A.E. & Goldsmith, N., 1998 - Early Miocene area - Gaia 8: 1-92 rodents and lagomorphs from Israel - Journal of Thomas, H., Sen, S., Khan, M., Battail, B. & Ligabue, Vertebrate Paleontology, Abstracts: 87 A G., 1982 - The lower Miocene fauna of Al-Sarrar Wu, W.Y., Ye, J., Bi, S.-D. & Meng, J., 2000 - The disco- (Eastern province, Saudi Arabia) – ‘Atlal’ Journal of very of Late Oligocene dormice from China - Saudi Arabian Archeology 5 (3): 109-136 Vertebrata PalAsiatica 38 (1): 39-42 Tobien, H., 1968 - Paläontologische Ausgrabungen nach Jungtertiären Wirbeltieren auf der Insel Chios Received 19 May 2001 (Griechenlands und bei Maragheh (NW Iran) - Accepted 4 November 2002

DEINSEA - ANNUAL OF THE NATURAL HISTORY MUSEUM ROTTERDAM P.O.Box 23452, NL-3001 KL Rotterdam The Netherlands

418