The Miocene Mammal Record of the Vallès-Penedès Basin

Our reference: PALEVO 882 P-authorquery-v13

AUTHOR QUERY FORM

Journal: PALEVO Please e-mail your responses and any corrections to:

Article Number: 882 E-mail: [email protected]

Dear Author,

Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen annotation in the PDF file) or compile them in a separate list. Note: if you opt to annotate the file with software other than Adobe Reader then please also highlight the appropriate place in the PDF file. To ensure fast publication of your paper please return your corrections within 48 hours.

For correction or revision of any artwork, please consult http://www.elsevier.com/artworkinstructions.

Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by ﬂags in the proof. Click on the ‘Q’ link to go to the location in the proof.

Location in Query / Remark: click on the Q link to go article Please insert your reply or correction at the corresponding line in the proof Q1 Please conﬁrm that given names and surnames have been identiﬁed correctly. Q2 We have replaced ref [Agust«õ et al., 1989] by [Agust«õ et al., 1990] in the text. Please check and correct if necessary. Q3 Citation (Madern et al., this volume; de Bruijn et al., 1996) have not been found in the reference list. Please supply full details for these references. Q4 Please check the symbol [] in Table 1, and correct if necassary.

Please check this box or indicate your approval if you have no corrections to make to the PDF ﬁle

Thank you for your assistance. G Model PALEVO 882 1–21 ARTICLE IN PRESS

C. R. Palevol xxx (2015) xxx–xxx

1 Contents lists available at ScienceDirect Comptes Rendus Palevol

www.sciencedirect.com

General palaeontology

2 The miocene mammal record of the Vallès-Penedès basin

3 (Catalonia)

4 Le registre miocène des mammifères du bassin de Vallès-Penedès

5 (Catalogne)

a,∗ a,b a c,d 6 Q1 Isaac Casanovas-Vilar , Anneke Madern , David M. Alba , Lluís Cabrera , e b a 7 Israel García-Paredes , Lars W. van den Hoek Ostende , Daniel DeMiguel , a,f a g,a c,d 8 Josep M. Robles , Marc Furió , Jan van Dam , Miguel Garcés , a h 9 Chiara Angelone , Salvador Moyà-Solà

10 a Institut Català de Paleontologia Miquel Crusafont, Universitat Autònoma de Barcelona, Edifici ICTA-ICP, Carrer de les Columnes s/n, 11 Campus de la UAB, 08193 Cerdanyola del Vallès, Barcelona, Spain 12 b Naturalis Biodiversity Center, PO Box 9517, 2300 RA Leiden, the Netherlands 13 c Departament d’Estratigrafia, Paleontologia i Geociències Marines, Facultat de Geologia, Universitat de Barcelona, Martí i Franqués s/n, 14 08028 Barcelona, Spain 15 d Institut Geomodels, Grup de Recerca Consolidat de Geodinàmica i Anàlisis de Conques, Universitat de Barcelona, Martí i Franqués s/n, 16 08028 Barcelona, Spain 17 e Departamento de Paleontología, Facultad de Ciencias Geológicas, Universidad Complutense de Madrid, Instituto de Geociencias IGEO 18 (CSIC, Universidad Complutense de Madrid), José Antonio Novais 2, 28040 Madrid, Spain 19 f FOSSILIA Serveis Paleontològics i Geològics, S.L., Carrer Jaume I 87, 1er 5a, 08470 Sant Celoni, Barcelona, Spain 20 g Department of Earth Sciences, Utrecht University, Budapestlaan 4, 3584 CD Utrecht, the Netherlands 21 h ICREA at Institut Català de Paleontologia Miquel Crusafont and Unitat d’Antropologia Biològica (Department BAVE), Universitat 22 Autonòma de Barcelona, Edifici ICTA-ICP, Carrer de les columnes s/n, Campus de la UAB, 08193 Cerdanyola del Vallès, Barcelona, Spain

24 article info abstract

25 26 Article history: The land mammal record of the Vallès-Penedès Basin (Catalonia, NE Spain) ranges from the 27 Received 16 March 2015 Early Miocene (Ramblian) to the Late Miocene (Turolian), that is from about 20 to 7 Ma. 28 Accepted after revision 27 July 2015 Here we present an updated review of the mammal succession focusing on biochronology 29 Available online xxx as well as on environmental and faunal changes. Based on faunal similarities with central Europe, we interpret this basin as a transitional zone between the forested environments of 30 Handled by Larry Flynn northern regions and the more arid landscapes of the inner Iberian Peninsula. The quality 31 of the Vallès-Penedès record and its chronostratigraphic control is clearly better for the 32 Keywords: Late Aragonian and the Vallesian (between 12.6–9.0 Ma), especially for small mammals. 33 Iberian Peninsula Therefore, we analyze small mammal diversity dynamics during this interval. Contrary to 34 Western Europe previous analyses, which found an abrupt extinction event coinciding with the Early/Late 35 Miocene 36 Magnetostratigraphy Vallesian boundary (the Vallesian Crisis), our results show that this pattern is due to uneven 37 Biostratigraphy sampling. Instead, taxonomic richness slowly decreased since the Late Vallesian as a result 38 Paleodiversity of a series of extinctions that mostly affected forest-dwelling taxa. © 2015 Académie des sciences. Published by Elsevier Masson SAS. All rights reserved.

∗ Corresponding author. E-mail address: [email protected] (I. Casanovas-Vilar).

Please cite this article in press as:Casanovas-Vilar, I., et al., The miocene mammal record of the Vallès-Penedès basin (Catalonia). C. R. Palevol (2015), http://dx.doi.org/10.1016/j.crpv.2015.07.004 G Model PALEVO 882 1–21 ARTICLE IN PRESS

2 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx résumé

39 40 Mots clés : Le registre fossile des mammifères du bassin de Vallès-Penedès (Catalogne, Nord-Est de 41 Péninsule Ibérique l’Espagne) s’étend du Miocène inférieur (Ramblien) jusqu’au Miocène supérieur (Turolien), 42 Europe de l’Ouest soit environ de 20 à 7 Ma. Ici nous présentons une révision actualisée des successions 43 Miocène de mammifères, en prêtant attention aux aspects biochronologiques, environnementaux 44 Magnétostratigraphie et faunistiques. Sur la base des similarités observées avec les faunes d’Europe centrale, 45 Biostratigraphie nous interprétons ce bassin comme une zone transitionnelle entre, d’une part, les envi- 46 Paléodiversité ronnements boisés des régions nordiques, et, d’autre part, comme le paysage plus aride de la partie intérieure de la Péninsule Ibérique. La qualité du registre fossile du Vallès- Penedès, ainsi que son contrôle chronostratigraphique, sont nettement supérieurs pour la partie Aragonienne supérieure et Vallésienne (entre 12,6–9,0 Ma), particulièrement pour les petits mammifères. Pour cette raison, nous analysons ici la dynamique de la diversité des petits mammifères pour cette période. À l’inverse des analyses précédentes, qui obser- vent une extinction brutale à la limite Vallésien inférieur/supérieur, nos résultats montrent que ce modèle résulte en fait d’un d’échantillonnage inégal. Nous proposons plutôt une diminution lente de la richesse taxonomique à partir du Vallésien supérieur, conséquence d’une série d’extinctions touchant, en priorité, les espèces forestières. © 2015 Académie des sciences. Publié par Elsevier Masson SAS. Tous droits réservés.

47 1. Introduction During the 1970s and the 1980s the knowledge of the 85 Vallès-Penedès mammal record improved with the addi- 86

48 The study of the mammal faunas of the Vallès-Penedès tion of new sites and reviews of important mammal groups 87 49 Basin (Barcelona, Catalonia) dates back to the late 19th such as the rodents, insectivores and artiodactyls (for syn- 88 50 century, when a few mammal remains were found in the thetic reviews, see Agustí et al., 1985 or Golpe-Posse, 1974). 89 51 short-lasting lignite mine of la Fontsanta (next to els Casots The publication of the first comprehensive paleoecologi- 90 52 site, CS in Fig. 1) and the Molí Calopa brickyard in Rubí cal and diversity analyses followed soon after (Agustí and 91 53 (MC, Figs. 1 and 2). These were reported by the priest Moyà-Solà, 1990; Agustí et al., 1984), resulting in the def- 92 54 and geologist Jaume Almera, from the Museum of Geol- inition of a remarkable extinction event during the early 93 55 ogy of Barcelona Seminary, who left the description of Late Miocene. The Vallesian Crisis, as it was called, was 94 56 these remains (most of them currently lost) in the hands also recognized in other European regions (Agustí and 95 57 of renowned mammal paleontologists of his time, includ- Moyà-Solà, 1990; Agustí et al., 1997, 1999; Fortelius and 96 58 ing M. Boule, C. Depéret and A. Gaudry (Almera, 1898). In Hokkanen, 2001; Fortelius et al., 1996). Finally, during 97 59 the following decades the outcrops of the Vallès-Penedès the last two decades, the construction of major works 98 60 were surveyed by paleontologists from the Barcelona Sem- such as highways and landfills has allowed the discov- 99 61 inary resulting in the discovery of new sites, including the ery of hundreds of new sites, mostly of Latest Aragonian 100 62 ones at els Hostalets de Pierola (Bataller, 1938). However, and Vallesian age. These new sites are placed in long and 101 63 the systematic sampling of the Miocene outcrops did not continuous series that have also been sampled for mag- 102 64 start until the 1940s being led by Miquel Crusafont, Josep F. netostratigraphy (Alba et al., 2012a; Garcés et al., 1996; 103 65 de Villalta and Jaume Truyols. This resulted in the discov- Moyà-Solà et al., 2009). The resulting correlations to the 104 66 ery of tens of new sites and the collection of thousands Geomagnetic Polarity Time Scale (GPTS) have allowed a 105 67 of specimens, most of which are currently housed at high precision in the age estimation of many sites, which, 106 68 the Institut de Paleontologia Miquel Crusafont (ICP) in coupled with the intensive sampling and study of the mam- 107 69 Sabadell. Crusafont and Villalta focused on the study of mal faunas, open the way for more refined paleoecological 108 70 mammals and occasionally other vertebrates (Crusafont and diversity analyses. Here we provide an updated review 109 71 et al., 1955; Crusafont Pairó, 1950; Crusafont Pairó and of the Miocene mammal record of the Vallès-Penedès and 110 72 de Villalta, 1951; de Villalta Comella and Crusafont Pairó, its geological context, together with an analysis of the 111 73 1941a, 1941b), whereas Truyols mostly devoted his stud- mammal diversity dynamics during the Late Aragonian and 112 74 ies to the stratigraphical context and regional geology the Vallesian, which is the best-sampled part of our record. 113 75 of the area (Crusafont Pairó and Truyols Santonja, 1954, 76 1960; Crusafont et al., 1955). Thanks to their work, the 2. Geological setting 114 77 Vallès-Penedès soon became a reference area for the 78 study of European Miocene faunas, and eventually the The northwestern Mediterranean was formed during 115 79 Vallesian land mammal age was defined based on its the latest Oligocene and the Miocene as a result of the sub- 116 80 record (Crusafont Pairó, 1950). The Vallesian age was duction of the old oceanic crust of the Tethys Ocean under 117 81 quickly accepted and used in other regions of the Old the Eurasian plate as the African plate progressively con- 118 82 World to characterize the late Miocene fossil faunas that verged with it (Roca and Guimerà, 1992; Roca et al., 1999). 119 83 are characterized by the entry of the first hipparionin At the Catalan Continental Margin, a widespread system 120 84 horses. of NE-SW and NNE-SSW oriented horst and half-grabens 121

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 3

Fig. 1. Simplified geological map of the Vallès-Penedès Basin and major Miocene mammal sites (modified from Institut Cartogràfic i Geològic de Catalunya, 2015). The area enclosed by the dashed perimeter is shown in detail in Fig. 2. For locality acronyms see Table 1. Fig. 1. Carte géologique simplifiée du bassin de Vallès-Penedès et principaux sites miocènes à mammifères (modifié d’après l’Institut Cartogràfic i Geològic de Catalunya, 2015). La zone délimitée par la ligne en pointillées est montrée en détail à la Fig. 2. Pour l’acronyme des localités, se référer au Tableau 1.

122 was created, making up the northwestern margin of the The Miocene record of the Vallès-Penedès Basin has 152 123 Mediterranean (Bartrina et al., 1992; Cabrera and Calvet, been split into four main lithostratigraphical units dated by 153

124 1996; Cabrera et al., 2004; Roca and Guimerà, 1992; Roca means of biostratigraphy and magnetostratigraphy (Agustí Q2 154 125 et al., 1999). The onshore zone of this continental mar- et al., 1985, 1990, 1997; Cabrera, 1981a, 1981b; Cabrera 155 126 gin includes the Vallès-Penedès and el Camp de Tarragona et al., 1991; Cabrera and Calvet, 1996; de Gibert and 156 127 half-grabens that are bounded by the horsts deﬁned by the Casanovas-Vilar, 2011; Garcés et al., 1996). The oldest of 157 128 Pre-littoral range to the northwest and the Littoral range these units is the Basal Breccia Unit, which crops out at a 158 129 to the SE (Garraf-Montnegre horst). few points in the Littoral Range. It consists of time trans- 159 130 The Vallès-Penedès Basin (Fig. 1) is an elongated half- gressive monogenic conglomerates and breccias. In the 160 131 graben of about 100 km long by 12–14 km wide. Its Vallès sector this unit dates back to the Ramblian (MN3), 161 132 northwestern margin is bounded by the Vallès-Penedès whereas in the Penedès sector its oldest record is Early 162 133 master fault and its basement dips gently towards this mar- Aragonian (MN4) in age (de Gibert and Casanovas-Vilar, 163 134 gin where attains up to 4000 m in depth (Bartrina et al., 2011). 164 135 1992; Roca et al., 1999). Some major faults with a vertical The overlying Lower Continental Units (Figs. 1 and 3) 165 136 slip larger than 1000 m, associated with minor faults with crop out mainly near the southeastern basin margin and 166 137 up to a few hundred meters of slip, occur at this margin. All consist of intensely red alluvial fan and alluvial-lacustrine 167 138 these southeastern margin faults were overlapped by Early deposits corresponding to the early Miocene (Ramblian- 168 139 to Middle Miocene sedimentary sequences, which were Early Aragonian, MN3-MN4). Their oldest reported sed- 169 140 affected by later minor fault reactivations (Cabrera, 1981a, imentary record corresponds to small alluvial-fan facies 170 141 1981b; Cabrera and Calvet, 1996). The sedimentary inﬁll sourced by local catchments from the southeastern basin 171 142 of the Vallès-Penedès Basin started in the Early Miocene margins. The overlying alluvial fan deposits covered wider 172 143 (Ramblian–Early Burdigalian; Cabrera, 1981a; Cabrera areas and were also sourced from the northwestern reliefs. 173 144 et al., 1991, 2004; de Gibert and Casanovas-Vilar, 2011) Shallow carbonate and evaporitic lacustrine systems devel- 174 145 and ended in the Late Miocene (Turolian–Tortonian). Major oped in zones in the southeastern part of the basin, such as 175 146 features of the stratigraphic record in the Vallès-Penedès the La Costablanca, Molí de Can Calopa and Sant Andreu 176 147 half-graben were closely controlled by the tectonic activ- de la Barca areas (Cabrera, 1981a, 1981b; Cabrera et al., 177 148 ity of its main bounding faults and the sea level changes in 1991). These lacustrine deposits are cyclically alternating 178 149 the western Mediterranean (Bartrina et al., 1992; Cabrera, with distal facies of alluvial fan systems, the latter bearing 179 150 1981b; Cabrera and Calvet, 1996; Cabrera et al., 2004; Roca most of the mammal and plant sites of this age (Agustí et al., 180 151 et al., 1999). 1985; Casanovas-Vilar et al., 2011a). 181

4 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx

Fig. 2. Detailed geological map of the western Vallès sector (see Fig. 1) showing the position of the main mammal sites (modiﬁed after Garcés et al., 1996). For locality acronyms see Table 1. Fig. 2. Carte géologique détaillée du secteur ouest de Vallès (voir Fig. 1) montrant la position des principaux sites à mammifères (modiﬁé d’après Garcés et al., 1996). Pour l’acronyme des localités, se référer aux Tableau 1.

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 5

Fig. 3. Paleogeographic schemes showing the evolution of the Vallès-Penedès Basin during late Oligocene and the Miocene: a: Chattian-Aquitanian (Late Oligocene-Earliest Miocene); b: Early Burdigalian; c: Langhian; d: Tortonian. Note the extension of shallow marine environments and reef systems during the Langhian (see also Fig. 1). Modiﬁed after Cabrera et al. (2004). Fig. 3. Présentation de l’évolution du bassin de Vallès-Penedès au cours de l’Oligocène supérieur et le Miocène : a : Chatien-Aquitanien (Oligocène supérieur-Miocène basal) ; b : Burdigalien basal ; c : Langhien ; d : Tortonien. Noter l’extension des environnements marins peu profonds et des récifs coralliens au cours du Langhien (voir aussi Fig. 1). Modiﬁé d’après Cabrera et al. (2004).

6 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx

182 During the latest Early Miocene and the Middle Miocene et al., 2012). However, sampling is uneven and most of the 240 183 (17–15 Ma) several sea level changes took place in the sites correspond to the Late Aragonian and Early Vallesian. 241 184 context of the Mid-Miocene Climatic Optimum (Zachos Regarding the localities, approximately two thirds of them 242 185 et al., 2001). The faults that bounded the southeastern basin have only small mammal remains, but a remarkable num- 243 186 margin became inactive and the southern structural cul- ber has provided both small and large mammals. A list of 244 187 minations were extensively onlapped and overlapped by the main sites and fossiliferous sections is given in Table 1. 245 188 the sedimentary infill (Cabrera and Calvet, 1996; Cabrera Finally, there are a few localities from which mostly large 246 189 et al., 1991; Roca et al., 1999). At least three episodes of mammal remains have been recovered. Most of the old col- 247 190 marine transgression and regression affected the Vallès- lections were assembled by the means of surface collecting 248 191 Penedès Late Burdigalian, Langhian and Early Serravalian; as a result of quarrying activity at brickyards, where the 249 192 Cabrera and Calvet, 1996; Cabrera et al., 1991; de Gibert and workers often found mammal fossils during manual dig- 250 193 Casanovas-Vilar, 2011; Roca et al., 1999). The sequences ging. The use of systematic collecting techniques increased 251 194 deposited at that time make up the Marine and Transi- since the late 1950s and early 1960s. These include planned 252 195 tional Units (Figs. 1 and 3). Since the southwestern basin excavations as well as systematic screen-washing of sed- 253 196 end and its southern margin in the Penedès were more iment samples of the major sites. During the early 1980s 254 197 directly connected and open to the sea, carbonate coral- screen-washing techniques were improved, with the intro- 255 198 gal platform deposits, marine bay and transitional fan-delta duction of sieving equipment capable of processing huge 256 199 siliciclastic systems occupied persistently that area. During sediment samples (Daams and Freudenthal, 1988). There- 257 200 Langhian times, the sea level was high enough and marine fore, larger micromammal samples, comprising between 258 201 environments extended up to the Vallès sector with devel- fifty to a few hundreds of individuals, could be gathered for 259 202 opment of shallow marine and transitional deposits. The many known sites. During the 1990s and the 2000s, most 260 203 Early Serravalian marine transgression only affected the of the mammal sites known to date were discovered as a 261 204 southwestern basin sectors. result of emergency campaigns coinciding with major con- 262 205 From Serravallian time onwards, sedimentation was struction porjects. Macrommamal remains were mostly 263 206 again dominated by alluvial fan sediments, which con- collected after systematic excavation, whereas screen- 264 207 stitute the Upper Continental Units (Figs. 1 and 2). The washing was used for the micromammals. Thus, most of 265 208 Vallès-Penedès main fault remained active until the Tor- the material has been obtained using standardized col- 266 209 tonian, thus controlling the evolution of Middle Aragonian lecting methods. The material is kept in the collections of 267 210 to middle Turolian alluvial fan systems sourced from the the Institut Català de Paleontologia Miquel Crusafont (ICP) 268 211 northwest reliefs (Fig. 3; Agustí et al., 1985, 1997; Cabrera at Sabadell, except for part of the old collections, which are 269 212 and Calvet, 1996; Casanovas-Vilar et al., 2008; de Gibert curated in the Geology Museum of the Barcelona Seminary 270 213 and Casanovas-Vilar, 2011; Garcés et al., 1996; Roca et al., and in the Museum of Natural Sciences of Barcelona. 271 214 1999). Thick, coarse-grained sequences were deposited 215 near the northwestern margin of the basin, whereas finer- 3.2. Magnetostratigraphy 272 216 grained, medial to distal-terminal sequences developed to 217 the south and southeast. Large alluvial fan systems (e.g., Magnetostratigraphic sampling in the Vallès-Penedès 273 218 Olesa-les Fonts fan) attained a radius of 10–15 km, while Basin started during the mid 1990s (Garcés et al., 1996) and 274 219 others were restricted to a few km of radial spread. The focused on the Vallesian successions. More recent studies 275 220 middle and distal areas of these fan systems interfered and have complemented these data with those from long sec- 276 221 coalesced with each other during their evolution. Most of tions covering the late Aragonian and its transition to the 277 222 the mammal sites of the Vallès-Penedès Basin are located in Vallesian (Alba et al., 2012a; Moyà-Solà et al., 2009). The 278 223 the distal to terminal, mudstone-dominated facies (Fig. 2). sequences studied for magnetostratigraphy include four 279 224 The Messinian event resulted in a major interrup- main sections: Abocador de Can Mata, Les Fonts, Mon- 280 225 tion of the sedimentation as the Mediterranean sea level tagut and Viladecavalls, plus a few shorter ones (Fig. 4). 281 226 experienced a major lowering (Cabrera and Calvet, 1996). These sequences contain abundant mudstone beds, which 282 227 Later on, during the early Pliocene, alluvial-fluvial (Gallart, allowed a tight sampling resolution of 1–2 m/site. Routine 283 228 1981) and shallow marine deposits (Martinell, 1988) were stepwise demagnetization of the natural remanence mag- 284 229 deposited over deeply entrenched erosive surfaces affect- netization revealed the presence of a stable magnetization 285 230 ing both the basement and Earlier Miocene sequences. carried by iron oxides such as magnetite and hematite. 286 231 Finally, Pleistocene-Holocene terraced alluvial and collu- Since magnetostratigraphy relies on the demonstrably pri- 287 232 vial sediments were deposited unconformably overlying mary character of the rock magnetization, a number of 288 233 Neogene series. stability tests (Opdyke and Channel, 1996) were performed. 289 In the Vallès-Penedès sequences, the relative uniformity 290 234 3. Materials and methods of bedding dip attitudes made the fold test insignificant, 291 but a northerly directed mean direction after tilt cor- 292 235 3.1. Collecting techniques rection suggests a pre-folding magnetization. A positive 293 conglomerate test further supported a primary, synsedi- 294 236 The Vallès-Penedès Miocene mammal record comprises mentary origin. More significantly, magnetostratigraphic 295 237 more than 300 sites ranging from the late Ramblian to the results could be cross-checked with biostratigraphy along 296 238 middle Turolian, that is from about 19.6 Ma (Larrasoana˜ multiple composite sections of Montagut, Les Fonts, Ter- 297 239 et al., 2006) to 7.4/6.8 Ma (Agustí et al., 2001; Hilgen rassa and Viladecavalls (Fig. 4), thus indicating that the 298

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 7

Table 1 Localities mentioned in the text and figures arranged alphabetically following the acronyms used in Figs. 1, 2 and 4. For each site we indicate its correlation Q4 to the local zones as well as to the MN units. In the case of long local sections that comprise several sites (ACM, ECM, EDAR) we give the acronym for the series but not individual acronyms for each site. In the case of the ACM series the whole biostratigraphic range covered by the series is given. Tableau 1 Localités mentionnées dans le texte, ainsi que les figures arrangées de faç on alphabétique, en référence aux acronymes utilisés pour les Fig. 1, 2 et 4. Pour chaque site, nous avons indiqué leur corrélation à la zone locale, ainsi qu’à l’unité MN. Dans le cas de coupes longues, incluant plusieurs sites (ACM, ECM, EDAR), nous donnons l’acronyme de la coupe, mais pas celui de chaque localité. Dans le cas de la coupe ACM, toute l’étendue biostratigraphique couvert par la série est donnée.

Acronym Site/Series Local zone Local subzone MN unit

ACM Abocador de Can Mata series Hispanomys M. crusafonti + D. MN6MN7+8 (includes more than 250 sites) assemblage lartetiD.crusafonti- Hippotherium interval B40OV Autovia Orbital de Barcelona Cricetulodon range C. hartenbergeri MN9 B40 Tram Olesa de Montserrat-Viladecavalls (includes 6 sites) CA Can Canals – – MN4 CAL Ca n’Almirall Hispanomys M. crusafonti + D. larteti MN6 assemblage CB Castell de Barberà Hispanomys HippotheriumC. hartenbergeri MN9 assemblage interval CBL la Costablanca – – MN3 CCN20 Creu Conill 20 Hispanomys HippotheriumC. hartenbergeri MN9 assemblage interval CCN22 Creu Conill 22 Hispanomys HippotheriumC. hartenbergeri MN9 assemblage interval CCT2 Camí de Can Tarumbot 2 Rotundomys R. montisrotundi MN10 abundance CF Can Sant Feliu (also known as Hispanomys Hippotherium-C. hartenbergeri MN9 Can Feliu) assemblage interval CFEU Can Feu (includes 2 sites) Cricetulodon range C. hartenbergeriP. hispanicus MN9 interval CJ Can Julià – – MN4 CL1 Can Llobateres 1 Cricetulodon range C. hartenbergeriP. hispanicus MN9 interval CL2 Can Llobateres 2 Cricetulodon range C. sabadellensis + P. hispanicus MN10 CM1 Can Mata 1 Hispanomys D.crusafonti–Hippotherium MN9 assemblage interval CM3 Can Mata 3 Hispanomys Hippotherium–C. hartenbergeri MN7+8 assemblage interval CMI Can Missert Hispanomys HippotheriumC. hartenbergeri MN9 assemblage interval CMV Can Martí Vell (includes 2 sites) – – MN4 CO les Conilleres Hispanomys M. crusafonti + D. larteti MN6 assemblage CP Can Poncic 1 Cricetulodon range C. hartenbergeri MN9 CPL Can Pallars de Llobateres Cricetulodon range C. hartenbergeriP. hispanicus MN9 (includes 2 sites) interval CPU Can Purull – – MN10 CRU Can Cruset (includes 2 sites) Rotundomys R. bressanus MN10 abundance CS els Casots – – MN4 CSB Can Casablanca Rotundomys R. montisrotundi MN10 abundance CT Can Turu Rotundomys R. bressanus MN10 abundance CV Ceràmiques Viladecavalls Rotundomys R. montisrotundi MN10 abundance EC el Canyet – – MN4 ECM Ecoparc de Can Mata series Hispanomys Hippotherium-C. hartenbergeri MN9 (includes 12 sites) assemblage interval EDAR Estació Depuradora d’Aigües Cricetulodon range C. hartenbergeri MN9 Residuals del Riu Ripoll (includes 15 sites) GO La Gornal – – MN6? LTR1 La Tarumba 1 Rotundomys R. montisrotundi MN10 abundance LV Les Cases de la Valenciana – – MN4 PA Les Escletxes del Papiol – – MN4 PO Polinyà (includes 4 sites) – – MN9

8 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx

Table 1 (Continued)

Acronym Site/Series Local zone Local subzone MN unit

RK11 Autopista de Rubí-Terrassa K11 Hispanomys D. crusafontiHippotherium MN7+8 assemblage interval ROS-A1 Ronda Oest de Sabadell A1 Zone K (Teruel – MN11 Basin) ROS-D6 Ronda Oest de Sabadell D6 Rotundomys R. bressanus MN10 abundance RR Riu Ripoll Rotundomys R. montisrotundi MN10 abundance RT10 Autopista de Rubí-Terrassa 10 Rotundomys R. montisrotundi MN10 abundance RT12 Autopista de Rubí-Terrassa 12 Cricetulodon range C. sabadellensis + P. hispanicus MN10 RT3B Autopista de Rubí-Terrassa 3B Cricetulodon range C. hartenbergeri RT3 C Autopista de Rubí-Terrassa 3 C Hispanomys M. crusafonti + D. larteti MN6 assemblage RT6 Autopista de Rubí-Terrassa 6 Cricetulodon range C. hartenbergeri MN9 (includes 4 sites) RT7 Autopista de Rubí-Terrassa 7 Cricetulodon range C. sabadellensis + P. hispanicus MN10 (includes 2 sites) RT8 Autopista de Rubí-Terrassa 8 Cricetulodon range C. sabadellensis + P. hispanicus MN10 SAB Sant Andreu de la Barca – – MN4 SM Sant Mamet – – MN4 SMT Sant Miquel de Toudell – – MN10 SPR Sant Pere de Ribes (includes – – MN6? the sites: Vinedos˜ and Pedrera MN7+8? C1) SQ Sant Quirze (includes several Hispanomys D. crusafontiHippotherium MN7+8 classical sites plus 3 assemblage interval microvertebrate sites) TF Torrent de Febulines (includes Rotundomys R. bressanus MN10 6 sites) abundance TFR Turó de les Forques (also – – MN3 known as Costablanca II) TNA Trinxera Nord Autopista Rotundomys R. bressanus MN10 (includes 2 sites) abundance TSA Trinxera Sud Autopista Rotundomys R. bressanus MN10 (includes 4 sites) abundance TT Torrentet de Traginers (also Zone L (Teruel – MN12 known as Piera) Basin) VI Vilobí – – MN4 VI20 Vilobí 20 – – MN5? MN6?

299 retrieved magnetic zonation bears chronostratigraphic sig- The calculation of sampling probability and generic 321 300 niﬁcance. richness requires the use of quantitative data. In this regard, 322 we used the maximum number of individuals of each taxon, 323

301 3.3. Calculation of sampling probability and taxonomic since this information is routinely written down in our col- 324 302 richness lection inventories. Sampling probability was calculated 325 using the method devised by Barry et al. (2002) as mod- 326

303 Time series diversity analyses require a great accuracy ified by Van der Meulen et al. (2005) and Van Dam et al. 327 304 in the dating of individual sites as well as a well-resolved (2006). This method considers the probability of finding a 328 329 305 taxonomy. Considering the quality of our data, diversity given taxon based on its abundance and sample size: 306 analyses can only be performed for the Late Aragonian and r 307 Early Vallesian, for which most of the sites have associ- p = 1 − (1 − q) 330 308 ated bio- and magnetostratigraphical data of high quality. 309 Regarding the taxonomic data, the quality of the record where q is the relative abundance of the taxon in the 331 310 is not the same for all mammal groups. Small mammals, interval where it was first (or last) recorded and r is the 332 311 and in particular the orders Rodentia and Eulipotyphla, are additional number of specimens sampled. A genus is con- 333 312 more abundant and have a more continuous record than sidered to be absent when the cumulative probability of 334 313 large mammals. Furthermore, they have been thoroughly finding it over preceding or succeeding time intervals is 335 314 reviewed and genus-level identifications are available greater than 0.9. Not surprisingly, the inferred range for a 336 315 (Casanovas-Vilar et al., research in progress). Between the given taxon may extend beyond its last observed record 337 316 Late Aragonian (12.6 Ma) and the end of the Vallesian and earlier than its first observed record. This is ultimately 338 317 (9.0 Ma) these orders are represented by more than 20,000 related to its abundance, so that rare taxa tend to have long 339 318 specimens from nearly 200 different sites. Our calculations inferred ranges unless sample size is large enough to reject 340 319 are carried out at the genus level, excluding all the material that they were present. We used a range-through assump- 341 320 that could not be identified to this taxonomic rank. tion, so we consider a genus to be continuously present 342

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 9

Fig. 4. Correlation of the key local magnetostratigraphical sections of the Vallès-Penedès Basin to the Geomagnetic Polarity Time Scale (GPTS). The European Land Mammal Ages (ELMA) and local biozones of the Vallès-Penedès are also indicated. We have indicated some of the major sites in the Can Mata composite series. Abbreviations in the Can Mata composite series refer to different sectors of the landﬁll and nearby farm houses as follows: ACM: Abocador de Can Mata [Can Mata landﬁll]; BCV: Barranc de Can Vila [Can Vila ravine]; BDL: Bassa de Decantació de Lixiviats [setting pond of leachates]; C1–C9: ACM cells1 to 9; CCV: Camí de Can Vila [Can Vila road]; CM: Can Mata. For other locality acronyms see Table 1. Neogene time scale and boundaries after Hilgen et al. (2012). GPTS after Ogg (2012). Fig. 4. Corrélation des coupes clés magnétostratigraphiques locales du bassin de Vallès-Penedès à la « Geomagnetic Polarity Time Scale » (GPTS). Les « European Land Mammal Ages » (ELMA), ainsi que les biozones locales de Vallès-Penedès, sont aussi indiqués. Nous avons indiqué quelques-uns des sites majeurs de la série composite de Can Mata. Les abréviations de la série composite de Can Mata font référence aux différents secteurs de la décharge et des fermes environnantes comme suit : ACM : Abocador de Can Mata [décharge de Can Mata] ; BCV : Barranc de Can Vila [ravin de Can Vila] ; BDL : Bassa de Decantació de Lixiviats [bassin de décantation de lixiviats] ; C1–C9 : ACM cellules 1 à 9 ; CCV : Camí de Can Vila [route de Can Vila] ; CM : Can Mata. Pour les autres acronymes des localités, voir Tableau 1. Échelle temporelle néogène et limites d’après Hilgen et al. (2012). GPTS d’après Ogg (2012).

10 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx

343 between its first and last appearances. This assumption polarity changes in the section that correspond to chron 401 402 344 is in contrast with Van der Meulen et al. (2005) who boundaries. Then age is calculated as follows: 345 interpreted discontinuous ranges as local extinctions fol- (LAge − UAge) 346 lowed by subsequent migrations. In our case only a few Age = UAge + · SD 403 347 very rare taxa show discontinuous ranges, so we rather T 348 attribute their absence to sampling biases. The studied time where Age is the estimated absolute age of the local- 404 349 span (12.6–9.0 Ma) was divided into 0.1 Myr bins, and ity; UAge and LAge are the ages of the upper and lower 405 350 observed and inferred genera ranges were used to calculate boundaries of the geomagnetic chron, respectively; T is the 406 351 generic richness for each bin (Fig. 7). Considering that the stratigraphic thickness of that magnetozone in the local 407 352 typical extent of time-averaging for terrestrial vertebrate section; and SD is the stratigraphic distance between the 408 353 assemblages in fluvial systems and surface accumulations locality and the top of the magnetozone in the same sec- 409 354 oscillates between 100 and 100,000 years (Behrensmeyer, tion. When one of the chron boundaries is not recorded 410 355 1982; Behrensmeyer et al., 2000), shorter time bins are not in the section, Age is calculated using the composite sec- 411 356 recommended. Finally, we have evaluated the effects of tions that do record the chron boundaries. This method 412 357 sample size in the calculation of these richness measures assumes constant sedimentation rates within each mag- 413 358 using the Spearman’s rank order correlation test. netozone, a condition which strictly speaking will not be 414 fulfilled because the sedimentation in alluvial fan systems 415 is episodic and there are short periods of erosion, lead- 416 359 4. Chronological framework and locality age ing to the existence of many short hiatuses (Sadler, 1981). 417 360 estimation However, if these hiatuses are randomly distributed within 418 the sections and their duration is not large compared to 419 361 The chronostratigraphic framework of the Vallès- the duration of an interval, this approach will give reliable 420 362 Penedès Basin is based on a combination of magneto- and results (Badgley et al., 1986; Barry et al., 2002; Johnson 421 363 biostratigraphic data. Unfortunately, magnetostratigraphic et al., 1988). In our study case, the relative stratigraphic 422 364 studies are still in progress for the older part of the record thickness of the corresponding geomagnetic intervals is 423 365 (Ramblian–Middle Aragonian), so for this interval the age of comparable between sections, implying similar sedimen- 424 366 the sites and main bioevents can only be constrained on the tation rates. Furthermore, no major hiatuses are known 425 367 basis of biostratigraphy. The same applies to the Turolian during the Late Aragonian and the Vallesian. 426 368 sites. In contrast, most of the late Aragonian and Valle- 369 sian sites are placed in the composite magnetostratigraphic 370 framework, which have a total thickness of approximately 5. The Miocene mammal succession: 427 371 700 m (Fig. 4). The overall record results in a long pattern biostratigraphy, main bioevents and 428 372 of polarity reversals, which allows a correlation with the paleoenvironmental reconstruction 429 373 GPTS (Ogg, 2012) using the unambiguous identification of 374 chron C5n, a distinctly long normal chron that character- Throughout the text, the Vallès-Penedès record is cor- 430 375 izes the early Tortonian as a reference point. This chron is related to the European Land Mammal Ages as well as to 431 376 recognized both in the Vallès sector (Garcés et al., 1996) the MN zonation. We have used the mammal ages that 432 377 and in the Can Mata composite series of els Hostalets de are customarely used for the Iberian record. This implies 433 378 Pierola (Moyà-Solà et al., 2009), in the Penedès sector. The some differences for the early and middle Miocene, since 434 379 overall magnetostratigraphic sequence represents a time the Aegenian, Orleanian and Astaracian ages are replaced 435 380 span of approximately 3.6 Myr, from 12.6 to 9.0 Ma (Fig. 4). by the Ramblian and the Aragonian ages (see Steininger, 436 381 Shorter sections, such as Terrassa or Can Llobateres, cannot 1999). For the internal subdivision and boundaries of these 437 382 be directly matched to the GPTS, but can be reliably cor- ages, the reader is referred to Daams and Freudenthal 438 383 related to longer sections on the basis of biostratigraphical (1989), Agustí et al. (2011) and Larrasoana˜ et al. (2006) 439 384 and magnetostratigraphical data. An absolute age up to the for the Ramblian, and to Daams et al. (1999), Agustí et al. 440 385 0.1 Myr bin could be assigned to most of these sites. Finally, (2011) and Van der Meulen et al. (2011, 2012) for the Arago- 441 386 approximately 35% of the sites (mostly older collections) nian. As far as MN zones are concerned, we follow the more 442 387 of this period lack associated magnetostratigraphical data. ‘biostratigraphical’ concept of the MN zones (see Hilgen 443 388 Therefore, their correlation relies entirely on local bios- et al., 2012; Van Dam, 2003) and use the boundaries for 444 389 tratigraphy, which usually implies an age uncertainty not Western Europe. The MN boundaries for Central Europe are 445 390 higher than 0.3 Ma (see below and Fig. 5). However, our significantly older, particularly for the Early Miocene; the 446 391 analyses require that each site is assigned an absolute age, reader is referred to Van der Meulen et al. (2011, 2012) 447 392 so a random age between its maximum and minimum pos- for a comparison. MN zone boundaries are based on dif- 448 393 sible ages is given to the sites with only biostratigraphical ferent sources: for the Ramblian, zone boundaries follow 449 394 data. Larrasoana˜ et al. (2006), Agustí et al. (2011) and Hilgen 450 395 Concerning the sites with associated magnetostrati- et al. (2012); for the Aragonian, they are mostly based in 451 396 graphical information, an absolute age can be estimated Daams et al. (1999), Agustí et al. (2011) and Van der Meulen 452 397 using the methods of Barry et al. (2002). The calculations et al. (2011, 2012); those of the Vallesian follow Agustí et al. 453 398 are based on the stratigraphic position of a given site in (2001) and Hilgen et al. (2012); and finally, for the Turolian 454 399 the local magnetostratigraphic series. The interpolation we follow Van Dam et al. (2001, 2006) and Hilgen et al. 455 400 must be done between two points of known ages, namely (2012). 456

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 11

Fig. 5. Local biozonation of the Vallès-Penedès Basin for the Late Aragonian to the Middle Turolian. The observed ranges of the biostratigraphically important taxa are indicated. The shadowed time intervals highlight the times of major faunal turnover: 1) Dispersal of hipparionin horses and some eastern faunal elements (Machairodus, Giraffidae) at the Aragonian/Vallesian boundary; 2) Vallesian Crisis, implying the disappearance of many forest elements of Middle Miocene origin and coinciding with the first record of murids (Progonomys hispanicus) and the cricetines of the genus Rotundomys; 3) Diversification of the murids and dispersal of the first ‘Pikermian’ elements such as the bovid Tragoportax and the hyaenid Adcrocuta. See main text for details. In the case of the MN zones, asterisks indicate the chronological position of their reference sites, whereas their boundaries are defined on the basis of selected appearance events (see Hilgen et al., 2012). Neogene time scale and boundaries after Hilgen et al. (2012). GPTS after Ogg (2012). Fig. 5. Biozonation locale du bassin de Vallès-Penedès de l’Aragonien supérieur au Turolien moyem. Les domaies stratigraphiques des taxons importants du point de vue biostratigraphique sont indiqués. Les intervalles de temps ombrés soulignent les changements fauniques majeurs : 1) Dispersion des chevaux hipparionins et des éléments fauniques orientales (Machairodus, Giraffidae) à la transition Aragonien/Vallésien ; 2) Crise Vallésienne, impliquant la disparition de nombreuses formes forestières trouvant leur origine au Miocène moyen, et synchroniquement, l’apparition des muridés (Progonomys hispanicus) et des cricétinés du genre Rotundomys ; 3) Diversification des muridés et dispersion des premiers éléments « pikermiens », comme par exemple le bovidé Tragoportax et le hyanidé Adcrocuta. Se reporter au texte pour détails. Pour ce qui concerne les zones MN, les astérisques indiquent la position chronologique des localités références, tandis que leurs limites sont définies sur la base de la sélection de premières apparitions de taxa sélectionnés. Échelle temporelle néogène et limites d’après Hilgen et al. (2012). GPTS d’après Ogg (2012).

457 5.1. The Ramblian (ca. 20.4–16.8/17 Ma) outcrop of the Basal Breccia Unit. This site contains scarce 472 macromammal remains and a richer rodent assemblage 473 458 The Early Miocene successions of the Vallès-Penedès that includes the cricetid Melissiodon, the eomyid Ligerimys 474 459 Basin have produced relatively few sites, and many of cf. fahlbuschi, the glirid Pseudodryomys ibericus, and scarce 475 460 them, particularly the older ones, are very poor. Crusafont remains of the cricetodontine Democricetodon. The ﬁrst 476 461 et al. (1955) systematically surveyed this part of the record common occurrence of Democricetodon marks the begin- 477 462 and found about ﬁfteen mammal-bearing sites, although ning of the Aragonian, but this genus is already recorded 478 463 many of them have since been destroyed (Casanovas-Vilar in the Latest Ramblian, though it is very rare (Van der 479 464 et al., 2011b). In recent years, some classical sites (such Meulen et al., 2012). Therefore, this site is correlated with 480 465 as Costablanca or Molí Calopa) have been surveyed again the Late Ramblian zone A, which according to the lat- 481 466 in order sample microvertebrates, which had been over- est data from the Ebro Basin would range between 19.6 482 467 looked by earlier researchers. The correlation of these and 16.8–17 Ma (Larrasoana˜ et al., 2006). The Ramblian is 483 468 localities is entirely based on biostratigraphy. Magne- chronologically equivalent to zone MN3 and its upper part 484 469 tostratigraphical studies are now in progress. corresponds to the so-called “cricetid vacuum” (Daams and 485 470 The oldest site in the Vallès-Penedès Basin is Turó de Freudenthal, 1989), that is, the time interval between the 486 471 les Forques (TFR, Figs. 1 and 2), which is located in a small last occurrence of the old Oligocene cricetids of the genus 487

12 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx

488 Eucricetodon and the earliest cricetodontines of the genus Valenciana, Can Martí Vell, and more importantly els Casots 547 489 Democricetodon. During this time Melissiodon, a peculiar (LV, CMV and CS, Fig. 1). Els Casots was excavated dur- 548 490 genus showing a unique “honeycomb” molar morphology, ing the 1990s and has produced a rich collection of micro- 549 491 is the only cricetid present in Western Europe. and macromammals in a remarkable state of preservation 550 492 The remaining Ramblian sites of the Vallès-Penedès (Casanovas-Vilar et al., 2011c). In the Vallès sector, the Early 551 493 show a similar rodent fauna dominated by the glirid gen- Aragonian sites are located in red lutites corresponding to 552 494 era Pseudodryomys and Simplomys, as well as eomyids of the distal facies of major fan systems. Some sites, such as 553 495 the genus Ligerimys. Relatively rich Ramblian macromam- les Escletxes del Papiol or Sant Mamet (PA and SM, Fig. 2) 554 496 mal faunas are known from Molí Calopa, Costablanca and are just a few meters below the Langhian marine units. The 555 497 Sant Andreu de la Barca (MC, CBL and SAB in Figs. 1 and 2). Vallès sites have mostly micromammals. 556 498 These include abundant remains of the minute artiodactyl The beginning of the Aragonian is marked by 557 499 Cainotherium miocaenicum, although the cervid Procervulus the first abundant occurrence of the cricetodontines 558 500 dichotomus is also common (Casanovas-Vilar et al., 2011a, Democricetodon and, slightly later, Megacricetodon (Daams 559 501 2011b). The equid Anchitherium and the felid Styriofelis are and Freudenthal, 1988, 1989; Daams et al., 1999; Van der 560 502 present at various sites, whereas the first gomphotheres Meulen et al., 2012). These cricetids originated in Asia 561 503 (Gomphotherium) are recorded at Sant Andreu de la Barca (Flynn and Wessels, 2013; Maridet et al., 2011) and soon 562 504 (Agustí and Galobart, 1997). Mein (1975a) used the earliest after their earliest record in Western Europe they became 563 505 record of proboscideans in Europe to characterize the MN4 major components of the Aragonian and Earliest Valle- 564 506 zone (16.8/17–16/16.4 Ma), but for Western Europe the sian faunas (e.g., Daams et al., 1988). In the Aragonian 565 507 first common occurrence of Democricetodon is now used as type area, the Calatayud-Montalbán Basin, the presence 566 508 diagnostic criterium for the base of this zone and the begin- of Megacricetodon distinguishes local zones B and C, both 567 509 ning of the Aragonian (Hilgen et al., 2012; Van der Meulen covering the Early Aragonian (Daams and Freudenthal, 568 510 et al., 2011, 2012). In the Vallès-Penedès (Sant Andreu de la 1988; Daams et al., 1999; Van der Meulen et al., 2012). 569 511 Barca) and France (Navère; Bulot and Ginsburg, 1993) the In the Vallès-Penedès, Democricetodon (D. hispanicus) and 570 512 record of proboscideans precedes that of Democricetodon, Megacricetodon (M. primitivus) coexist in all the stud- 571 513 and is therefore placed in the late MN3, with an age ran- ied localities but in Sant Mamet. However, the absence 572 514 ging from 19.6 to 16.8/17 Ma. This age is congruent with of Megacricetodon may be due to insufficient sampling. 573 515 recent dating of the proboscidean dispersal event in Cen- Furthermore, the paracricetodontine Eumyarion (E. wein- 574 516 tral Europe at ca. 17.5 Ma (Pálfy et al., 2007). However, the furteri) is also present in this site (Agustí, 1981), and its first 575 517 appearance of this order in other areas of Eurasia is sub- appearance in other Spanish basins is in zone C (Daams 576 518 stantially older, with deinotheres present at the Pakistan and Freudenthal, 1974; Ruiz-Sánchez et al., 2003; Van 577 519 Siwaliks at around 23 Ma (Flynn et al., 2013) and an uniden- der Meulen et al., 2012). Even though D. hispanicus and 578 520 tified elephantoid having been reported from the latest M. primitivus dominate the rodent assemblages in most 579 521 Oligocene of the Bugti Hills (Antoine et al., 2003, 2013). sites, the eomyid Ligerimys ellipticus and the glirids Pseu- 580 522 Macroplant remains are abundantly reported from the dodryomys ibericus and Simplomys simplicidens are common 581 523 shallow lacustrine levels of the Costablanca series and offer components as well (Agustí, 1981, 1983; Agustí and Llenas, 582 524 a detailed picture of the local paleoenvironment surround- 1993; Casanovas-Vilar et al., 2011c). None of the known 583 525 ing the humid areas (Sanz de Siria Catalán, 1993, 2001). sites in the Vallès-Penedès could be correlated to the 584 526 The recovered flora points to a tropical to subtropical cli- Lowermost Aragonian (zone B), where characteristically 585 527 mate with rainfall seasonality (Sanz de Siria Catalán, 1993, Democricetodon is the only cricetid genus present. Inter- 586 528 2001). This reconstruction is congruent with the small estingly, the Calatayud-Montalbán local zones could be 587 529 mammal fauna, which is dominated by glirids with a simple extended to the Ramblian and early Aragonian record of 588 530 dental pattern that purportedly preferred more arid envi- the Vallès-Penedès (Casanovas-Vilar et al., 2011a, 2011b), 589 531 ronments than most of their relatives (Van der Meulen and a situation contrasting with that of the Late Aragonian 590 532 De Bruijn, 1982). However, the large mammal fauna, which and especially the Vallesian, when the rodent succession 591 533 includes basal pecorans (Amphitragulus), cervids (Procervu- shows marked differences between these areas (see below; 592 534 lus) and small browsing equids (Anchitherium), is more Madern et al., this volume). Q3 593 535 consistent with forested environments (Casanovas-Vilar Most of our data on the early Aragonian large mam- 594 536 et al., 2011a). mal faunas of the Vallès-Penedès come from the site of 595 els Casots (Casanovas-Vilar et al., 2011c and references 596 537 5.2. The Early Aragonian (16.8/17–16/16.4 Ma) therein). The suid Eurolistriodon, the early bovid Eotragus 597 and the tragulid Dorcatherium appear for the first time 598 538 The Early Aragonian is chronologically equivalent to the (Agustí et al., 1985; Casanovas-Vilar et al., 2011a, 2011c). 599 539 MN4. The sites are located in facies of small lacustrine sys- At this time, deinotheres are first recorded, being scarcely 600 540 tems or in distal facies of alluvial fans (Figs. 1 and 2). Most represented at els Casots and les Escletxes del Papiol. 601 541 of the known localities were first reported by Crusafont Cainotherium persists, but is rarer than in earlier sites. 602 542 et al. (1955). The richest sites belong to the Subirats unit, The suoids (Eurolistriodon, Taucanamo), paleomerycids 603 543 in the Penedès sector, which is defined by an alternation (Ampelomeryx) and some rhinoceroses (Plesiaceratherium) 604 544 of lutites and carbonates that correspond to shallow lakes are common among the large mammal faunas. The most 605 545 and the distal facies of alluvial fans (Agustí and Cabrera, abundant carnivores are the amphicyonids (Amphicyon) 606 546 1980; Cabrera, 1981a, 1981b). These include les Cases de la and ursids (Hemicyon, Plithocyon). 607

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 13

608 5.3. The Middle Aragonian (16/16.4–13.8 Ma) represented by several different species during this time 666 interval. This allows defining a long Hispanomys Assem- 667 609 During the Langhian (15.97–13.65 Ma), great parts of blage Zone that covers the Late Aragonian and Earliest 668 610 the basin were covered by a shallow sea embayment, so it is Vallesian (c. 12.5–10.3 Ma, Fig. 5). The coexistence of dif- 669 611 not surprising that no mammal sites are correlated to the ferent species of Megacricetodon and Democricetodon,as 670 612 approximately time-equivalent Middle Aragonian, which well as the first record of the equid Hippotherium, allow 671 613 equals to MN5. However, some portions of the Vallès sec- the subdivision of the Hispanomys zone into four sub- 672 614 tor were exposed at that time, so they might potentially zones (Fig. 5): M. crusafonti + D. larteti Concurrent Range 673 615 yield sites of this age. On the other hand, there are numer- Subzone (?–12.4 Ma); M. crusafonti–D. crusafonti Interval 674 616 ous paleobotanical sites around the towns of Martorell and Subzone (12.4–11.9 Ma); D. crusafonti–Hippotherium Inter- 675 617 Sant Sadurní d’Anoia. These have produced a rich collection val Subzone (11.9–11.2 Ma); Hippotherium–Cricetulodon 676 618 of macroflora, indicating that the environment was slightly hartenbergeri Interval Subzone (11.2–10.3 Ma). The last cor- 677 619 more arid than during the Ramblian (Sanz de Siria Catalán, responds to the Earliest Vallesian, whereas the previous 678 620 1993, 2001). In addition, palynological samples have been three correlate to the Late Aragonian. Even though the 679 621 collected from marine sediments close to the small reef faunal succession is not the same, these subzones can be 680 622 systems of Sant Pau d’Ordal, located just a few kilometers correlated with the local zones of the Calatayud-Montalbán 681 623 to the south of Sant Sadurní d’Anoia (Bessedik and Cabrera, Basin (Daams et al., 1999). Subzone G2 from Calatayud- 682 624 1985). The pollen assemblage indicates the presence of Avi- Montalbán would be approximately equivalent to the M. 683 625 cennia mangroves in the coastline. crusafonti + D. larteti subzone, whereas the subzone G3 684 would be equivalent to the M. crusafonti–D. crusafonti and 685 626 5.4. The Late Aragonian (13.8–11.2 Ma) the D. crusafonti–Hippotherium interval subzones in the 686 Vallès-Penedès. Similarly, our zones can be correlated to 687 627 The Late Aragonian is one of the best represented the MN units, with the three late Aragonian subzones being 688 628 time intervals in the Vallès-Penedès Basin. Most of our equivalent to the MN 7 + 8 (Hilgen et al., 2012). Neverthe- 689 629 knowledge of this period derives from the study of the less, if one follows the definition of the MN zones of Agustí 690 630 intensively-sampled Can Mata composite series at els et al. (2001) or Mein (1999), the older subzone (M. crusa- 691 631 Hostalets de Pierola (Penedès sector; Fig. 1). This series fonti + D. larteti) would be equivalent to uppermost part of 692 632 (Fig. 4) comprises the sections of the Abocador de Can Mata the MN6. 693 633 landfill (ACM), the Ecoparc de Can Mata (ECM), and a few As explained above, the Late Aragonian small mam- 694 634 nearby classical sites (Can Mata 1 and 3), which together mal faunas are dominated by Hispanomys, Democricetodon 695 635 include more than 200 mammal sites that have produced and Megacricetodon. The paracricetodontine Eumyarion 696 636 tens of thousands of specimens (Alba et al., 2011). The Can may be very common at certain sites, such as ACM/BCV1 697 637 Mata composite series ranges from about 12.5 Ma to 10.6 (see Fig. 4). The glirids are diverse though not abun- 698 638 Ma, the age of the sites being well constrained thanks to dant and include mostly purported forest-dwelling genera 699 639 a combination of detailed litho-, bio- and magnetostrati- (Bransatoglis, Muscardinus, Glirudinus, Myoglis). The small 700 640 graphical data (Fig. 4; Alba et al., 2011; Casanovas-Vilar eomyids Keramidomys and Eomyops as well as the fly- 701 641 et al., 2011d; Moyà-Solà et al., 2009). In addition to the Can ing squirrels (Albanensia, Miopetaurista) are first recorded 702 642 Mata series, there are a number of other sites that have in the M. crusafonti + D. larteti subzone, ca. 12.5 Ma. Cas- 703 643 been correlated to the late Aragonian. These include the torids (Chalicomys) are recorded ca. 11.9 Ma, but are very 704 644 classical Sant Quirze sites (SQ, Fig. 2), Ca n’Almirall and les rare. The hypsodont cricetid Anomalomys, appears for the 705 645 Conilleres (CAL and CO, Fig. 1; Agustí et al., 1985). The latter first time at ca. 11.8 Ma. The eulipotyphlan faunas are 706 646 two are placed in the upper transitional facies of the Marine diverse (e.g., Furió et al., 2011, 2015) and include talpids 707 647 and Transitional Unit and in palustrine facies related to flu- (Desmanella, Talpa), dimylids (Plesiodimylus), heterosori- 708 648 vial sequences in the Penedès sector. Agustí et al. (1985) cids (Dinosorex), abundant erinaceids (Galerix, Parasorex) 709 649 reported the presence of the cricetids Cricetodon aff. jotae, and crocidosoricine soricids. 710 650 Eumyarion medium and Megacricetodon crusafonti, which The large mammal faunas are diverse (Alba et al., 711 651 would indicate a correlation to MN6. However, the fauna 2011). The megaherbivores include two different gen- 712 652 has not been studied in detail, so other correlations can- era of proboscideans (Gomphotherium, Deinotherium), 713 653 not be ruled out. On the other hand, certain sites that had rare chalicotheres (Anisodon) and diverse rhinos 714 654 been traditionally placed in the late Aragonian, namely Can (Alicornops, Hoploaceratherium, Lartetotherium). The suoids 715 655 Missert and Castell de Barberà (CM and CB, Fig. 2), are here (Albanohyus, Conohyus, Listriodon, Propotamochoerus) are 716 656 correlated to the early Vallesian because scarce hippari- also diverse and common, whereas the cervid Euprox 717 657 onin remains were identified during a recent review of the may be common at some sites. There is a remarkable 718 658 collections (Robles et al., 2011; Rotgers and Alba, 2011). diversity of carnivorans, including mustelids, hyaenids, 719 659 A high-resolution biozonation for the Late Arago- felids (Pseudaelurus, Styriofelis) and false sabertooths 720 660 nian and Vallesian has been devised on the basis of (the barbourofelid Albanosmilus). The suid Listriodon 721 661 the rodent fauna (Casanovas-Vilar et al., 2011d and splendens, the small and short-legged rhino Alicornops 722 662 research in progress). The Late Aragonian rodent assem- and specially the moschid Micromeryx are the most 723 663 blages are characterized by the dominance of the cricetids abundant large mammals in the Can Mata composite 724 664 Hispanomys, Democricetodon and Megacricetodon. Usu- series. Hominoids (Anoiapithecus brevirostris) are first 725 665 ally, Hispanomys is the most abundant taxon, being recorded at around 12.3–12.4 Ma (Alba et al., 2013). This 726

14 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx

727 group shows a remarkable diversity around 11.9–12 Ma taxonomic level of genera and species. The Pikermian fauna 786 728 (M. crusafonti–D. crusafonti Interval Subzone), being expanded its range into Western Europe during the Turo- 787 729 represented by the genera Dryopithecus, Anoiapithecus lian (Bernor, 1983, 1984; Eronen et al., 2009). However, 788 730 and Pierolapithecus (Casanovas-Vilar et al., 2011d). Plio- Crusafont Pairó (1950) showed that some ‘Pikermian’ taxa 789 731 pithecoids first occur slightly later, at around 12 Ma such as the earliest hipparionins dispersed earlier. Cur- 790 732 (Alba et al., 2012b), but are mostly recorded between rently this dispersal is placed within chron C5r.1n in the 791 733 11.8–11.7 Ma (D. crusafonti–Hippotherium Interval Sub- Vallès-Penedès Basin, with an estimated age of 11.2 Ma 792 734 zone), being represented by the genus Pliopithecus. The (Garcés et al., 1996, 1997; dates recalculated taking into 793 735 cervid Euprox furcatus, a common component of the account the latest version of the GPTS after Ogg, 2012). This 794 736 fauna, is first recorded during the M. crusafonti + D. larteti age is not inconsistent with the record of the Calatayud- 795 737 subzone. Other ruminant groups, such as the bovids and Montalbán Basin, where the entry could be placed between 796 738 the tragulids, are generally rare. Amongst the bovids, 11.2 and 10.8 Ma (Garcés et al., 2003; Van Dam et al., 797 739 Miotragocerus and Tethytragus first occur during the D. 2014). It is also congruent with the data from Central 798 740 crusafonti–Hippotherium Interval Subzone. Tethytragus,an Europe (Bernor et al., 1988; Woodburne, 2009), but it is 799 741 abundant genus in the inner Iberian basins, is restricted to somewhat older than age estimates for this event in other 800 742 this zone, being represented by scarce remains only. areas (Sinap Formation, Siwaliks), which range from 10.8 801 743 Overall, the Late Aragonian faunas of the Vallès-Penedès to 10.3 Ma (Barry and Flynn, 1989; Barry et al., 1982, 1985, 802 744 Basin stand out by their high diversity due to the presence 2002; Flynn et al., 2013; Kappelman et al., 2003). The old- 803 745 of a number of taxa that are extremely rare or not recorded est occurrence of hipparionins (Hippotherium) is recorded 804 746 at all in other Iberian Basins (Agustí, 1989; Casanovas-Vilar at the lowermost part of the Montagut composite sec- 805 747 and Agustí, 2007; Casanovas-Vilar et al., 2008; DeMiguel tion (Fig. 4). However, the Aragonian/Vallesian transition 806 748 et al., 2011). These taxa are traditionally recognized as is documented in the Can Mata composite series, the sites 807 749 forest dwellers, shared by the Vallès-Penedès and other belonging to the ECM section and the classical locality of 808 750 European regions such as France and Germany, but absent Can Mata 3 (CM3) being correlated to the Vallesian (Fig. 4). 809 751 from inner Spain. Amongst the small mammals these CM3 has numerous hipparionin remains, which would be 810 752 include certain eulipotyphlans (dimylids, talpids), arboreal slightly younger than the ones of the Montagut section. On 811 753 glirids, pteromyinins, eomyids, and certain cricetids such the other hand, these equids are not recorded in any of the 812 754 as Eumyarion and Anomalomys. The large mammal faunas ECM sites. The Earliest Vallesian rodent faunas cannot be 813 755 support these inferences, since certain taxa such as homi- distinguished from those of the latest Aragonian and are 814 756 noids, pliopithecoids or chalicotheres are not recorded placed within the Hispanomys Assemblage Zone (Fig. 5). 815 757 from elsewhere in the Iberian Peninsula. Others, such as Only the presence of the equid Hippotherium confirms their 816 758 suoids and rhinocerotids, are more abundant and usually Vallesian age. The Hippotherium–Cricetulodon hartenberg- 817 759 more diverse in the Catalan basins. Unfortunately, there are eri Interval Subzone defines the earliest Vallesian. In some 818 760 no time-equivalent paleobotanical data that could provide rodent faunas, the cricetodontine Megacricetodon ibericus 819 761 a more complete picture of the late Aragonian paleoenvi- may be extremely abundant, such as Creu Conill 22 or Can 820 762 ronments in the basin. Feliu (CCN22, CF, Fig. 2). 821 The rest of the Vallesian subzones are defined on the 822 763 5.5. The Early Vallesian (11.2–9.73 Ma) basis of the evolution of the cricetine genera Cricetu- 823 lodon and Rotundomys. The species present are members 824 764 The Vallesian mammal successions have been the sub- of a well-known anagenetic lineage (C. hartenbergeri–C. 825 765 ject of multiple reviews, given that the Vallès-Penedès is sabadellensis–R. montisrotundi–R. bressanus), character- 826 766 the type area for this mammal age (Agustí and Moyà- ized by its increasing hypsodonty coupled with a 827 767 Solà, 1990; Agustí et al., 1985, 1997; Casanovas-Vilar et al., simplification of the molar pattern into a continuous sig- 828 768 2011a, 2011d; Moyà-Solà and Agustí, 1987, 1989). Most of moidal crest (Freudenthal, 1967; Freudenthal et al., 1998; 829 769 the Vallesian sites are placed near the towns of Sabadell Mein, 1975b). The Cricetulodon Range Zone (Fig. 5) cov- 830 770 and Terrassa, in the Vallès sector of the basin (Fig. 2), and ers most of the Early Vallesian (MN9) and the beginning 831 771 have associated magnetostratigraphical data (Fig. 4; Agustí of the Late Vallesian (MN10). This zone is divided into 832 772 et al., 1997; Garcés et al., 1996). three subzones depending on the Cricetulodon species 833 773 Crusafont Pairó (1950) defined the Vallesian on the basis present: C. hartenbergeri Range Subzone (10.3–10 Ma); 834 774 of the coexistence of the first hipparionin horses with fau- C. hartenbergeri–Progonomys hispanicus Interval Subzone 835 775 nal elements that were characteristic of the forest faunas of (10–9.7 Ma); and the short-lasting C. sabadellensis + P. 836 776 the middle Miocene, such as certain rhinos, cervids, suoids hispanicus Concurrent Range Subzone (9.7–9.6 Ma). Crice- 837 777 and chalicotheres. Previously, hipparionins were thought tulodon is the most common element in these subzones 838 778 to be characteristic of the so-called Pikermian faunas, after and it may define up to 70% of the recovered remains. The 839 779 the Turolian site of Pikermi, near Athens (Greece). The Pik- genera Democricetodon and Megacricetodon are rare and 840 780 ermian faunas are dominated by open country herbivores, only represented by the small-sized species. The castorids 841 781 such as antelopes, giraffes and horses, whereas forest ele- (Chalicomys, Euroxenomys) and the paracricetodontine 842 782 ments, such as deer, are rare (Bernor et al., 1996; Eronen Eumyarion are common at certain sites such as Can Llo- 843 783 et al., 2009; Solounias et al., 1999). Such faunas represent bateres 1 (CL1, Fig. 2). The micromammal faunas of this 844 784 a rather uniform mammal community from the standpoint subzone are diverse in forest elements, particularly glirids. 845 785 of their ecological characteristics, not necessaritly at the Eulipotyphlans reach their highest diversity, with at least 846

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 15

847 three genera of erinaceids, two genera of dimylids and late Vallesian assemblages of other Iberian basins, such as 906 848 talpids, and four of soricids (Van den Hoek Ostende the Teruel Basin, where murids are dominant and define 907 849 and Furió, 2005). The hitherto persisting heterosoricid the basis for local biostratigraphy (Van Dam, 1997; Van 908 850 Dinosorex is still a common element, but suddenly dis- Dam et al., 2001). By contrast, the genus Rotundomys 909 851 appears from the micromammal assemblage at the very is the most abundant component of the faunas in the 910 852 beginning of the Vallesian (Furió et al., 2015). The first Vallès-Penedès, defining the Rotundomys Abundance Zone 911 853 occurrence of murids in Western Europe defines the that covers most of the Late Vallesian (Fig. 5). This is 912 854 boundary between early and late Vallesian, or between divided into two subzones according to the member of 913 855 MN9 and MN10 (Agustí et al., 1997, 2001; Hilgen et al., the anagenetic lineage present: R. montisrotundi Lineage 914 856 2012). Progonomys hispanicus is the earliest murid in the Subzone (9.6–9.4 Ma) and the R. bressanus Lineage Sub- 915 857 Vallès-Penedès, being recorded at several sites in the zone (9.4–8.9 Ma). As the upper boundary for this zone 916 858 uppermost part of chron C4Ar.3r (Figs. 4 and 5), with an we take the Vallesian/Turolian age boundary in the Teruel 917 859 estimated age of 9.7 Ma. This rodent family evolved in the Basin (Krijgsman et al., 1996; Van Dam et al., 2001), since it 918 860 middle Miocene in Pakistan (Jacobs, 1977), while the genus is not recorded in any of the magnetostratigraphical sec- 919 861 Progonomys first appeared in that area around 12 Ma (Barry tions of the Vallès-Penedès. The rodent faunas are less 920 862 and Flynn, 1989; Barry et al., 1982, 2002; Flynn et al., diverse than during the Early Vallesian, although a num- 921 863 1995, 2013; Wessels, 2009). Updated stratigraphic corre- ber of forest elements are present in low numbers (Agustí 922 864 lations attained in the Calatayud-Montalbán Basin (Van et al., 1997, 1999; Casanovas-Vilar et al., 2011a, 2014). In 923 865 Dam et al., 2014) show a very rare presence of Progono- the R. montisrotundi subzone, Progonomys hispanicus is the 924 866 mys during the early Vallesian, at 10.4 Ma, and a common only murid species present, whereas in the R. bressanus 925 867 occurrence of this form after 10.0–9.9 Ma. In Turkey, its subzone the murid assemblage is enriched with the appear- 926 868 first record is slightly older, at 10.1 Ma (Kappelman et al., ance of P. cathalai and P. woelferi as well as the genera 927 869 2003). In our study area, the dispersal of Progonomys coin- Parapodemus and Huerzelerimys. Rotundomys montisrotundi 928 870 cides with the first appearance of Rotundomys (R. freirensis), coexists for some time with its putative descendant, R. 929 871 although C. sabadellensis is still the dominant rodent in bressanus. The large mammal faunas record a number of 930 872 most assemblages. At the same time, a number of taxa are new eastern immigrants, particularly during the R. bres- 931 873 no longer recorded, including the cricetids Democricetodon, sanus subzone (Agustí et al., 1997; Casanovas-Vilar et al., 932 874 Megacricetodon, Eumyarion and the glirids Microdyromys, 2011a, 2011d). These include the suids Schizoporcus and 933 875 Glirudinus, Bransatoglis and Muscardinus vallesiensis. Cas- Microstonyx, the equid Cremohipparion and the bovid Trago- 934 876 torids and pteromyines become rarer. This set of local portax, which will later become major components of the 935 877 extinctions is part of the Vallesian Crisis (see below). Turolian faunas. Amongst the carnivorans, the hyaenid 936 878 Regarding the large mammals, the Early Vallesian Adcrocuta is first recorded within the R. montisrotundi sub- 937 879 faunas retain large numbers of late Aragonian components, zone, whereas the ursid Indarctos is last recorded at the 938 880 so that the immigration of Hippotherium did not coin- same time. Finally, hominoid primates (Hispanopithecus) 939 881 cide with any noticeable extinction event (Agustí et al., are last recorded within this subzone, at the site of La 940 882 1997). Giraffids, the felid Machairodus and the bovid Aus- Tarumba 1 (LTR1, Fig. 2), which has an estimated age of 941 883 troportax immigrated at the same time (Casanovas-Vilar about 9.6 Ma (Casanovas-Vilar et al., 2011d). Pliopithecoids 942 884 et al., 2011a; Fig. 5). The long-ranging felid Pseudaelu- survived into the R. bressanus subzone, being last recorded 943 885 rus is last recorded within the C. hartenbergeri subzone. at 9.1 Ma at Torrent de Febulines (TF, Fig. 2; Moyà-Solà et al., 944 886 The hominoid Hispanopithecus is recorded at several sites 2001). 945 887 from the Cricetulodon zone (Casanovas-Vilar et al., 2011d). The Vallesian biozonation of the Vallès-Penedès can 946 888 Overall, the early Vallesian faunas are extremely diverse, be recognized in other Catalan basins, such as La Seu 947 889 including a high diversity of medium to large-sized herbi- d’Urgell or the Empordà basins (Casanovas-Vilar et al., 948 890 vores and carnivoran species (Agustí et al., 1997, 2001; 2011b). However, it is different from that in other areas 949 891 Casanovas-Vilar et al., 2011a). Bovids became more diverse of Spain, such as the Calatayud-Montalbán Basin (Daams 950 892 and abundant. This assemblage is very similar to that of and Freudenthal, 1988; Daams et al., 1999) or the Teruel 951 893 contemporaneous sites from Central Europe (Casanovas- basins (Van Dam et al., 2001; Van de Weerd, 1976). While 952 894 Vilar et al., 2008, 2010), such as Rudabánya in Hungary the order and timing of the main bioevents, such as the 953 895 (Bernor et al., 2005), which on the other hand has also pro- first record of Cricetulodon and Progonomys, is roughly the 954 896 duced great ape remains. Finally, the early/late Vallesian same, the major components of the faunas that are the basis 955 897 transition witnessed the extinction of several large mam- for the local biozonations differ. For example, Rotundomys 956 898 mal taxa of middle Miocene origin in the context of the is not recorded in Teruel, and in the Vallès-Penedès the 957 899 Vallesian Crisis (see below). These include the barbourofe- murids are too rare to be considered for the definition of 958 900 lid and amphycionid carnivorans, the suids Listriodon and local biozones. Casanovas-Vilar and Agustí (2007) empha- 959 901 Parachleuastochoerus, and the bovid Miotragocerus. sized these differences and related them to the occurrence 960 of more humid and forested environments in the Vallès- 961 902 5.6. The late Vallesian (9.7–8.9 Ma) Penedès, particularly during the late Vallesian. Such an 962 interpretation is supported by the distribution of other 963 903 Even though the murids are continuously present in mammal taxa such as certain glirids, the flying squirrels, 964 904 the Vallès-Penedès Basin, they remain usually rare com- chalicotheres and hominoid primates, to name just a few 965 905 ponents of the mammal faunas. This contrasts with the examples. 966

16 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx

967 Regarding the Vallesian paleoenvironment, the diverse 968 mammal fauna, rich in forest elements, has been inter- 969 preted as indicative of subtropical to warm-temperate 970 humid forest environments (Agustí and Moyà-Solà, 1990; 971 Agustí et al., 1984, 1999; Casanovas-Vilar and Agustí, 2007; 972 Morales et al., 1999). Unfortunately, the paleobotanical 973 information of this period is very scarce, and has to be com- 974 plemented with data from other Catalan basins (Empordà, 975 Cerdanya). These show that the zonal vegetation consisted 976 of a warm-temperate mixed forest defined by evergreen 977 laurels, together with leguminous trees and shrubs as well 978 as a significant proportion of deciduous elements (Sanz de 979 Siria Catalán, 1993, 1994). The subtropical taxa would have 980 been mostly restricted to the humid areas in the lowlands 981 (Marmi et al., 2012). The rich late Vallesian macroflora from 982 Trinxera Sud Autopista 2 near Terrassa (TSA2, Fig. 2) indi-

983 cates the presence of a warm-temperate mixed forest with Fig. 6. Composition of the large mammal faunas from the late Aragonian 984 up to 40% of deciduous taxa (Agustí et al., 2003). to the Middle Turolian. n stands for the maximum number of individuals identified to the genus level. The Late Aragonian faunas are represented by the Sant Quirze sites; the Early Vallesian includes Creu Conill 20, Can 985 5.7. The Turolian (8.9–5.3 Ma) Missert, Castell de Barberà, Can Poncic, Santiga and Can Llobateres 1; the Late Vallesian includes Trinxera Sud Autopista, Trinxera Nord Autopista 986 Only two Turolian sites are well known from the and Torrent de Febulines; finally, the middle Turolian is solely represented 987 Vallès-Penedès: Ronda Oest de Sabadell A1 in Sabadell by Torrentet de Traginers. Note the evenness of the Late Aragonian faunas, 988 and Torrentet de Traginers in Piera (ROS-A1 and TT, the abundance of equids during the Vallesian, and the dominance of bovids during the Middle Turolian. See text for further details. 989 Figs. 1 and 2). However, some fossiliferous areas, particu- Fig. 6. Composition des faunes de grands mammifères de l’Aragonien 990 larly around Viladecavalls and Piera, might potentially yield tardif au Turolien moyen. n correspond au nombre maximum d’individus 991 new Turolian sites. Furthermore, the Vallesian/Turolian identifiés au niveau générique. Les faunes aragoniennes tardives sont doc- 992 transition could be recorded at the Ronda Oest de Sabadell umentées par les sites de Sant Quirze ; le Vallésien précoce comprend Creu Conill 20, Can Missert, Castell de Barberà, Can Poncic, Santiga et Can 993 succession (Fig. 2), although these faunas and their strati- Llobateres 1 ; le Vallésien tardif comprend Trinxera Sud Autopista, Trinx- 994 graphical context have not been studied in detail yet. era Nord Autopista et Torrent de Febulines ; enfin, le Turolien moyen 995 ROS-A1 has a relatively rich small mammal sample dom- est seulement représenté par Torrentet de Traginers. À noter l’égalité des 996 inated by the murid Huerzelerimys vireti, which would faunes de l’Aragonien tardif, l’abondance des équidés durant le Vallésien, 997 indicate an Early Turolian (MN11 or zone K) age (Agustí et la dominance des bovidés durant le Turolien moyen. Se reporter au texte pour complément d’information. 998 et al., 2001; Van Dam et al., 2001, 2006). The large mammals 999 are insufficiently known, but the faunas are significantly 1000 impoverished with bovids and Hippotherium dominating the environment was clearly more humid and forested in 1026 1001 the assemblage. In contrast, Torrentet de Traginers (also the Vallès-Penedès Basin. Furthermore, although biostrati- 1027 1002 known in the literature as Piera) has produced a rich large graphical data are preliminary, the Teruel rodent-based 1028 1003 mammal fauna that includes more than 2700 specimens. local zonation can apparently be extended to the Vallès- 1029 1004 Its diversity is remarkably low, far below that of the Valle- Penedès Basin. 1030 1005 sian assemblages. The fauna is overwhelmingly dominated 1006 by the bovid Tragoportax gaudryi, with the equid Cre- 6. Late Miocene diversity dynamics and a 1031 1007 mohipparion mediterraneum and the giraffid Birgerbohlinia re-evaluation of the Vallesian Crisis 1032 1008 schaubi as major components (Fig. 6). The remaining taxa 1009 include the suid Microstonyx and the hyaenid Adcrocuta, Agustí and Moyà-Solà (1990) were the first to recognize 1033 1010 amongst others. Forest elements are extremely rare, but the Vallesian Crisis, an important turnover coinciding with 1034 1011 chalicotheres (Anisodon) and rhinoceroses (Aceratherium, the early/late Vallesian boundary (around 9.7 Ma), which 1035 1012 Dihoplus) are still present. Intensive screen-washing has implied the abrupt extinction of most of the forest-adapted 1036 1013 not been carried out at this site, so the small mammal sam- elements that had characterized the Middle Miocene. These 1037 1014 ple size is very limited, consisting of just twenty molars authors had previously recognized that some rodent and 1038 1015 belonging to the murid Occitanomys adroveri, which indi- artiodactyl taxa seemed to be missing during the late Valle- 1039 1016 cates a middle Turolian (MN12 or zone L) age (Agustí et al., sian (Agustí, 1981; Agustí et al., 1984, 1985; Moyà-Solà 1040 1017 2001; Van Dam et al., 2001, 2006). This contrasts with pre- and Agustí, 1987, 1989), but later found that other mam- 1041 1018 vious correlations, which mostly on the basis of the large mals, including certain carnivorans and the primates had 1042 1019 mammal fauna had correlated this site to MN11 (Agustí disappeared at the same time. The event was initially recog- 1043 1020 et al., 1985, 1997; Casanovas-Vilar et al., 2011a). The com- nized in the Vallès-Penedès Basin, although they suggested 1044 1021 position and structure of the mammal fauna is very similar that it had also occurred in other European regions (Agustí 1045 1022 to that of contemporaneous sites of the Teruel Basin (Alcalá, and Moyà-Solà, 1990). Later studies supported this conclu- 1046 1023 1994; Alcalá et al., 2000), pointing to a similar arid and sion, leading many authors to consider the Vallesian Crisis 1047 1024 warm environment in both regions. This is in sharp con- as a European or even Eurasian event (Agustí et al., 1997, 1048 1025 trast with the situation seen during the Vallesian, when 1999; Fortelius and Hokkanen, 2001; Fortelius et al., 1996). 1049

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 17

1050 However, the potential causes of the Vallesian Crisis are 1051 elusive, since the main climatic (Zachos et al., 2001) 1052 and vegetational changes (Mossbrugger et al., 2005) had 1053 already started during the Middle Miocene, almost six mil- 1054 lion years earlier. Therefore, the crisis has been related 1055 to a complex interplay of climatic factors which would 1056 have increased the temperature latitudinal gradient along 1057 Europe and, in particular, rainfall seasonality in the peri- 1058 Mediterranean area (Agustí et al., 1997, 1999, 2003). In 1059 the Vallès-Penedès, the crisis would have been triggered 1060 by an increase in temperature seasonality that implied a 1061 significant increase of deciduous trees in the zonal vege- 1062 tation (Agustí et al., 2003). However, deciduous elements 1063 were already major components of the floral assemblages 1064 in Central Europe (Mossbrugger et al., 2005) and other 1065 areas of Catalonia and the Iberian Peninsula (Barrón et al., 1066 2010). More recently, Agustí et al. (2013) admitted the 1067 lack of correlation of the Vallesian Crisis with any major 1068 climatic perturbations and proposed a rather ad hoc new Fig. 7. Small mammal (Rodentia and Eulipotyphla) observed generic 1069 model calling for internal causes. The Vallesian faunas were richness, sample size (maximum number of individuals) and richness 1070 highly diverse because of the entry of new immigrants and measures for the late Aragonian and the Vallesian. All the parameters 1071 the persistence of most Middle Miocene faunas. According are calculated for 0.1 Myr bins and a range-through approach is used for 1072 to these authors, this resulted in a sort of self-organized all richness measures. The light gray areas indicate gaps in the record. Note that ‘observed richness’ fluctuates importantly with sample size 1073 climax assemblage that was highly sensitive to environ- and is strongly correlated with it. ‘Range-through richness’ is based on 1074 mental changes, so that just small perturbations could raw specimen counts but uses a range-through approach (rs = 0.864; P 1075 make the whole system collapse like a ‘House of Cards’. [no correlation] = 2.21 × 10−11). Nevertheless, it is also highly correlated

1076 However, it has been recently shown that previous eval- with sample size (rs = 0.418; P [no correlation] = 0.012). Therefore, note 1077 uations of the Vallesian Crisis in the Vallès-Penedès had that the decrease in richness after the Vallesian Crisis is explained by the smaller sample size as compared to the Early Vallesian. ‘Inferred-range 1078 not taken into account the effects of uneven sampling on richness’ estimates consider the probability of recording a given taxon 1079 diversity estimates (Casanovas-Vilar et al., 2014). Here we given sampling effort and are uncorrelated with sample size (rs = 0.191; 1080 present an updated summary of those analyses that incor- P [no correlation] = 0.270). Inferred richness shows a gradual decrease 1081 porates the latest data from the Late Aragonian sites of during the Late Vallesian instead of an abrupt crisis. See text for further details. 1082 the Can Mata composite series as well as some Vallesian Fig. 7. Richesse générique observée, ainsi que la taille des échantillons 1083 sites that had not been included in previous work (Fig. 7). (nombre maximum d’individus) et les mesures d’abondance des petits 1084 Only rodents and eulipotyphlans are taken into account. mammifères aragoniens supérieurs et vallésiens. Tous les paramètres sont 1085 However, rodents in particular are considered to be highly calculés pour des intervalles de 0,1 M.a. et une approche « range-through » 1086 affected by the Vallesian Crisis (Agustí and Moyà-Solà, est utilisée pour toutes les mesures d’abondance. Les zones gris clair indiquent des manques dans l’enregistrement. À noter que l’ « abon- 1087 1990; Agustí et al., 1997, 1999). For the sake of compari- dance observée » fluctue fortement en fonction de la taille de l’échantillon 1088 son, we have included sample size in Fig. 7. The quality of et est aussi fortement corrélée avec elle. L’ « abondance range-through 1089 the record is clearly better during the Late Aragonian and » est basée sur un comptage des spécimens mais utilise une approche −11 1090 the Early Vallesian, with sample size decreasing markedly « range-through » (rs = 0,864 ; p [no correlation] = 2,21 × 10 ). Cepen- dant, elle est aussi hautement corrélée à la taille de l’échantillon (r = 0,418 1091 during the Late Vallesian. It is important to note that there s ; p [no correlation] = 0,012). Par conséquent, il est à noter que la baisse de la 1092 is a sampling peak just before the Early/Late Vallesian richesse générique observée après la crise vallésienne est expliquée par la 1093 boundary, thus immediately preceding the crisis. Genus taille plus réduite des échantillons, comparée aux données fournies pour 1094 richness tracks sample size and both variables are signif- le Vallésien inférieur. Les adeptes de l’« Abondance inférée au domaine » 1095 icantly correlated (rs = 0.418; P [no correlation] = 0.012), stratigraphique considèrent la probabilité de répertorier un taxon donné selon un effort d’échantillonnage donné et qu’il n’y a pas corrélation avec la 1096 even if a range-through approach is used as in Agustí taille de l’échantillon (rs = 0,191 ; p [no correlation] = 0,270). L’abondance 1097 et al. (2013). Therefore, the decrease in richness during inférée montre une baisse graduelle au cours du Vallésien terminal plutôt 1098 the late Vallesian is explained by the poorer quality of qu’une crise abrupte. Voir le texte pour plus de détails. 1099 the record. Furthermore, the sampling peak in the time 1100 interval before the early/late Vallesian boundary also helps 1101 to distort the pattern. As data simulations have shown 1102 (Foote, 2000), quality peaks will artificially inflate diver- The ‘victims’ of the ‘Vallesian Crisis’ have two things 1111 1103 sity and overestimate extinction rates because many taxa in common: they are mostly purported forest-dwellers 1112 1104 that would have last occurred in a preceding time interval of Middle Miocene origin; and, perhaps more impor- 1113 1105 are then last recorded in the interval with the better record. tantly, they are generally rare. Amongst the rodents, they 1114 1106 When we use inferred-range diversity estimates, which are mostly include the glirids, which, although diverse in many 1115 1107 uncorrelated with sample size (rs = 0.191; P [no correla- sites, rarely account for more than 10% of the recov- 1116 1108 tion] = 0.270), the Vallesian Crisis changes from an abrupt ered specimens. The same can be said of certain sciurids, 1117 1109 extinction event to a gradual decrease in genus richness such as the pteromyines, beavers, and the paracricetodon- 1118 1110 during the Late Vallesian (Fig. 7). tine Eumyarion, which are only common at certain sites. 1119

18 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx

1120 Therefore, sample size during the late Vallesian would not 7. Summary and conclusions 1181 1121 be sufficient to record these rare taxa. This is further sup- 1122 ported by our data from a few richer Late Vallesian sites, After more than seventy years of systematic survey, the 1182 1123 such as Torrent de Febulines. Some of the ‘Vallesian Cri- mammal record of the Vallès-Penedès Basin stands out 1183 1124 sis’ ‘victims’ still have been recovered at this site, including as one of the best Miocene records in Europe, compris- 1184 1125 a diverse dormice fauna (Bransatoglis, Myoglis, Glirudi- ing hundreds of sites and tens of thousands of specimens. 1185 1126 nus, Paraglirulus), flying squirrels (Miopetaurista), beavers It includes both large and small mammals and ranges 1186 1127 (Chalicomys) and the eomyid Eomyops. Not surprisingly, from the Early Miocene (Ramblian) to the Late Miocene 1187 1128 pliopithecoids, represented by the endemic genus Egarap- (Middle Turolian). However, the quality of the record is 1188 1129 ithecus, are also present at Torrent de Febulines (Moyà-Solà uneven and the late Aragonian and Vallesian successions 1189 1130 et al., 2001). However, some local disappearances seem to have been more intensely sampled during the last years. 1190 1131 have occurred during the early/late Vallesian transition, Moreover, a magnetostratigraphical study of the major 1191 1132 including those of the cricetids Eumyarion, Democricetodon mammal-bearing successions has been carried out and 1192 1133 and Megacricetodon, the small castorid Euroxenomys and hence accurate age estimates are available for many of 1193 1134 the hetererosoricid Dinosorex. these sites. During most of the Miocene, the Vallès-Penedès 1194 1135 The large mammal record of the Vallès-Penedès is not represented a transitional area between the Iberian and 1195 1136 as continuous as that of small mammals and, further- Central European mammal faunas, so that many of the 1196 1137 more, a taxonomic review of certain groups should be forest-adapted taxa present in the latter also occur in this 1197 1138 carried out before a similar diversity analysis is attempted. basin. By contrast, the faunas of other Spanish basins are 1198 1139 However, we can already see that the faunal composi- indicative of more arid environments that apparently pre- 1199 1140 tion of the Vallesian and, especially, the Turolian large vented the dispersal of some taxa, such as hominoids, 1200 1141 mammal assemblages differs markedly from that of the chalicotheres or certain rodents and insectivores, into inner 1201 1142 Late Aragonian (Fig. 6). In general, the Late Aragonian Spain. 1202 1143 assemblages consist of an even mixture of rhinocerotids, As a case study, we have analyzed small mammal diver- 1203 1144 suoids, cervids and moschids. During the Vallesian, the sity dynamics during the Late Aragonian and the Vallesian 1204 1145 bovids increase their abundance and hipparionin horses (from 12.6 to 9.0 Ma), which represents our best-sampled 1205 1146 become dominant components of the faunas. Finally, the interval. Previous studies had recognized a major extinc- 1206 1147 Turolian assemblages (solely represented by Torrentet de tion event, the Vallesian Crisis, characterized by the abrupt 1207 1148 Traginers) are the least diverse and are overwhelmingly disappearance of most forest-adapted taxa at the Early/Late 1208 1149 dominated by the bovid Tragoportax, followed by the equid Vallesian boundary (9.7 Ma). We show that previous diver- 1209 1150 Cremohipparion and the giraffid Birgerbohlinia as major sity analyses were greatly influenced by sampling, so that 1210 1151 components. The remaining taxa, which include some the decrease in diversity during the Late Vallesian is corre- 1211 1152 forest elements, are very rare. The dominance of savanna- lated with a decrease in the quality of the record. Clearly, 1212 1153 like herbivores is characteristic of the Turolian faunas, sample size is too poor to record the rare taxa, which 1213 1154 but the earlier change in evenness seen in the Vallesian mostly turn out to be purported forest dwellers. Neverthe- 1214 1155 assemblages may indicate that some of the changes that less, diversity slowly decreased during the late Vallesian as 1215 1156 would later shape the Turolian had already started in the a part of a more gradual extinction event, likely extend- 1216 1157 Vallesian. ing into the Turolian. Such analyses cannot be applied to 1217 1158 As we have seen, the uneven quality of the record the large mammal faunas yet, although the known Turo- 1218 1159 overemphasizes the severity and abruptness of the ‘Valle- lian assemblages show a lower diversity than the Vallesian 1219 1160 sian Crisis’ in the Vallès-Penedès Basin. Some extinction ones, being clearly dominated by a few taxa. Our future goal 1220 1161 did occur, but apparently they were merely part of a more consists of expanding the temporal and taxonomic range of 1221 1162 protracted turnover event extending into the Turolian. these analyses, in order to better understand faunal dynam- 1222 1163 Therefore, it is not surprising that the crisis did not coincide ics in this region. 1223 1164 with any remarkable climatic event at 9.7 Ma. The analysis 1165 of the geographical range of the ‘Vallesian Crisis’ is beyond Acknowledgements 1224 1166 the scope of this paper. However, we must note that, 1167 although some disappearances occur, one discrete event We specially would like to thank the organizers and the 1225 1168 is not recognized in other Iberian basins, such as the Teruel host committee of the XII Annual Meeting of the Euro- 1226 1169 Basin, although rodent extinction rates are higher between pean Association of Vertebrate Paleontologists (Torino, 1227 1170 10–9.5 Ma in this area (Van Dam et al., 2006). In Central June 2014) for their support. We also acknowledge the con- 1228 1171 Europe, some forest-adapted taxa (such as chalicotheres, structive comments and remarks by the contributors to the 1229 1172 moschids and certain rodents) survived well into the early symposium “Windows into deep time: dynamics of faunal 1230 1173 Turolian, being recorded in some Austrian (Kohfidisch, change in long continental successions” on our preliminary 1231 1174 Eichkogel) and German sites (Dorn-Dürkheim 1; Franzen results. This work has been supported by the Spanish Mini- 1232 1175 et al., 2013). Finally, the small mammal record of the sterio de Economía y Competitividad (CGL2011-28681, 1233 1176 eastern Mediterranean (Greece, Turkey) does not record CGL2010-1749 and RYC-2013-12470 to ICV), the Agèn- 1234 1177 significant extinctions (de Bruijn et al., 1996) and indeed cia de Gestió d’Ajuts Universitaris i de Recerca (2014 SGR 1235 1178 large mammal diversity increased continuously during the 416 and 2014 SGR 467) and the Departament de Cul- 1236 1179 Vallesian, peaking at the early Turolian (Fortelius et al., tura (2014/100584 and 2014/100609) of the Generalitat 1237 1180 1996). de Catalunya. IG-P acknowledges a JAE-Doc contract (CSIC) 1238

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 19

1239 co-funded by the European Social Fund and Research group remains of Anoiapithecus and the first appearance datum of hominoids 1312 in the Iberian Peninsula. J. Hum. Evol. 65, 573–584. 1313 1240 UCM-910607 “Evolution of Cenozoic Mammals and Conti- Alcalá, L., 1994. Macromamíferos neógenos de la fosa de Alfambra-Teruel. 1314 1241 nental Paleoenvironments”. AM acknowledges the support Instituto de Estudios Turolenses - Museo Nacional de Ciencias Natu- 1315 1242 of an Erasmus Staff Training mobility grant by Erasmus+ rales (CSIC), Madrid (554 p). 1316 1243 (Life Long Learning Programme). We deeply appreciate the Alcalá, L., Alonso-Zarza, A.M., Álvarez Sierra, M.A., Azanza, B., Calvo, I.P., 1317 Canaveras,˜ J.P., Van Dam, J.A., Garcés, M., Krijgsman, W., Van der 1318 1244 comments and constructive critics by the two reviewers Meulen, A.J., Morales, J., Peláez-Campomanes, P., Pérez González, A., 1319 1245 of this manuscript, J.C. Barry and R.L. Bernor, as well as by Sánchez Moral, S., Sancho, R., Sanz Rubio, E., 2000. El registro sedi- 1320 1246 the guest editor, L.J. Flynn. We thank Jérôme Prieto (Lud- mentario y faunístico de las cuencas de Calatayud-Daroca y Teruel. 1321 Evolución paleoambiental y paleoclimática durante el Neógeno. Rev. 1322 1247 wig Maximilians University Munich) for his help with the Soc. Geol. Espana˜ 13, 323–343. 1323 1248 French translation of the abstract and figure captions. Almera, J., 1898. Sobre la serie de mamíferos fósiles descubiertos en 1324 Cataluna.˜ Mem. Real Acad. Cien. Art.e Barcelona 2, 351–357. 1325 Antoine, P.-O., Welcomme, J.-L., Marivaux, L., Baloch, I., Benammi, M., 1326 Tassy, P., 2003. First record of Paleogene Elephantoidea (Mammalia, 1327 1249 References Proboscidea) from the Bugti Hills of Pakistan. J. Vert. Paleontol. 23, 1328 978–981. 1329 1250 Agustí, J., 1981. Roedores Miomorfos del Neógeno de Cataluna.˜ Ph. D. Antoine, P.-O., Métais, G., Orliac, M.J., Crochet, J.-Y., Flynn, L.J., Marivaux, L., 1330 1251 Dissertation. Universitat de Barcelona (288 p, 3 pl). Rahim Rajpar, A., Roohi, G., Welcomme, J.-L., 2013. Mammalian Neo- 1331 1252 Agustí, J., 1983. Roedores (Mammalia) del Mioceno inferior de Can Martí gene biostratigraphy of the Sulaiman Province, Pakistan. In: Wang, X., 1332 1253 Vell (Vallès-Penedès, Cataluna,˜ Espana).˜ Estudios Geol. 39, 417–430. Flynn, L.J., Fortelius, M. (Eds.), Fossil Mammals of Asia. Neogene Bios- 1333 1254 Agustí, J., 1989. The Miocene rodent succession in eastern Spain: a zoo- tratigraphy and Chronology. Columbia University Press, New York, pp. 1334 1255 geographical appraisal. In: Lindsay, E.H., Fahlbusch, V., Mein, P. (Eds.), 400–422. 1335 1256 European Neogene Mammal Chronology. Plenum Press, New York, pp. Badgley, C., Tauxe, L., Bookstein, F.L., 1986. Estimating the error of age 1336 1257 375–404. interpolation in sedimentary rocks. Nature 319, 139–141. 1337 1258 Agustí, J., Cabrera, L., 1980. Nuevos datos sobre la biozonación del Burdi- Barrón, E., Rivas-Carballo, R., Postigo-Mijarra, J.M., Alcalde-Olivares, C., 1338 1259 galiense continental de la cuenca del Vallès-Penedès. Acta Geol. Hisp. Vieira, M., Castro, L., Pais, J., Valle-Hernández, M., 2010. The Ceno- 1339 1260 3, 81–84. zoic vegetation of the Iberian Peninsula: a synthesis. Rev. Palaeobot. 1340 1261 Agustí, J., Galobart, A., 1997. Noves localitats amb mamífers fòssils en el Palynol. 162, 382–402. 1341 1262 miocè de la Conca del Vallès-Penedès. Trib. Arqueol. 1996–1997, 9–22. Barry, J.C., Flynn, L.J., 1989. Key biostratigrapic events in the Siwalik 1342 1263 Agustí, J., Llenas, M., 1993. Los roedores del Mioceno inferior de Els Casots sequence. In: Lindsay, E.H., Fahlbusch, V., Mein, P. (Eds.), Euro- 1343 1264 (Vallès-Penedès). Nota preliminar. In: Comunicaciones de las IX Jor- pean Neogene Mammal Chronology. Plenum Press, New York, 1344 1265 nadas de Paleontología. Universidad de Málaga, Málaga, pp. 70–72. pp. 557–571. 1345 1266 Agustí, J., Moyà-Solà, S., 1990. Mammal extinctions in the Vallesian (Upper Barry, J.C., Lindsay, E.H., Jacobs, L.L., 1982. A biostratigraphic zonation 1346 1267 Miocene). Lect. Not. Earth Sci. 30, 425–432. of the middle and upper Siwaliks of the Potwar Plateau of northern 1347 1268 Agustí, J., Moyà-Solà, S., Gibert, J., 1984. Mammal distribution dynamics Pakistan. Palaeogeogr. Palaeoclimatol. Palaeoecol. 37, 95–130. 1348 1269 in the eastern margin of the Iberian Peninsula during the Miocene. Barry, J.C., Johnson, N.M., Mahmood Raza, S., Jacobs, L.L., 1985. Neogene 1349 1270 Paleobiol. Cont. 14, 33–46. mammalian change in southern Asia: correlations with climatic, tec- 1350 1271 Agustí, J., Cabrera, L., Moyà-Solà, S., 1985. Sinopsis estratigráfica del tonic, and eustatic events. Geology 13, 637–640. 1351 1272 Neógeno de la fosa del Vallès-Penedès. Paleontol. Evol. 18, 57–81. Barry, J.C., Morgan, M.E., Flynn, L.J., Pilbeam, D., Behrensmeyer, A.K., Raza, 1352 1273 Agustí, J., Cabrera, L., Domènech, R., Matinell, J., Moyà-Solà, S., Ortí, F., S.M., Khan, I.A., Badgley, C., Hicks, J., Kelley, J., 2002. Faunal and envi- 1353 1274 de Porta, J., 1990. Neogene of the Penedès area (Prelittoral Catalan ronmental change in the Late Miocene Siwaliks of northern Pakistan. 1354 1275 Depression). Paleontol. Evol. 2, 187–207. Paleobiology 28 (3), 1–71. 1355 1276 Agustí, J., Cabrera, L., Garcés, M., Parés, J.M., 1997. The Vallesian mammal Bartrina, M.T., Cabrera, L., Jurado, M.J., Guimerà, J., Roca, E., 1992. Evo- 1356 1277 succession in the Vallès-Penedès basin (Northeast Spain): paleomag- lution of the central Catalan margin of the Valencia trough (western 1357 1278 netic calibration and correlation with global events. Palaeogeogr. Mediterranean). Tectonophysics 203, 219–247. 1358 1279 Palaeoclimatol. Palaeoecol. 133, 149–180. Bataller, J.R., 1938. Els ratadors fossils de Catalunya. Impremta de la Casa 1359 1280 Agustí, J., Cabrera, L., Garcés, M., Llenas, M., 1999. Mammal turnover and d’Assistència President Macià, Barcelona (64 p). 1360 1281 Global climate change in the Late Miocene terrestrial record of the Behrensmeyer, A.K., 1982. Time resolution in fluvial vertebrate assem- 1361 1282 Vallès-Penedès Basin (NE Spain). In: Agustí, J., Rook, L., Andrews, P. blages. Paleobiology 8, 211–227. 1362 1283 (Eds.), Hominoid Evolution and Climatic Change in Europe. Vol. 1: The Behrensmeyer, A.K., Kidwell, S.M., Gastaldo, R.A., 2000. Taphonomy and 1363 1284 Evolution of Neogene Terrestrial Ecosystems in Europe. Cambridge paleobiology. Paleobiology 26 (4), 103–147. 1364 1285 University Press, Cambridge, pp. 390–397. Bernor, R.L., 1983. Geochronology and zoogeographic relationships of 1365 1286 Agustí, J., Cabrera, L., Garcés, M., Krijgsman, W., Oms, O., Parés, J.M., 2001. Miocene Hominoidea. In: Ciochon, R.L., Corruccini, R. (Eds.), New 1366 1287 A calibrated mammal scale for the Neogene of western Europe. State Interpretations of Ape and Human Ancestry. Plenum Press, New York, 1367 1288 of the art. Earth Sci. Rev. 52, 247–260. pp. 21–64. 1368 1289 Agustí, J., Sanz de Siria, A., Garcés, M., 2003. Explaining the end of the Bernor, R.L., 1984. A zoogeographic theater and a biochronologic play: the 1369 1290 hominoid experiment in Europe. J. Hum. Evol. 45, 145–153. time/biofacies phenomena of Eurasian and African Miocene mammal 1370 1291 Agustí, J., Pérez-Rivarés, F.J., Cabera, L., Garcés, M., Pardo, G., Arenas, C., provinces. Paleobiol. Cont. 14, 121–142. 1371 1292 2011. The Ramblian-Aragonian boundary and its significance for the Bernor, R.L., Kovar-Eder, J., Lipscomb, D., Rögl, F., Sen, S., Tobien, H., 1988. 1372 1293 European Neogene continental chronology. Contributions from the Systematic, stratigraphic and paleoenvironmental contexts of first 1373 1294 Ebro Basin record (NE Spain). Geobios 44, 121–134. appering Hipparion in the Vienna Basin, Austria. J. Vert. Paleontol. 8, 1374 1295 Agustí, J., Cabrera, L., Garcés, M., 2013. The Vallesian mammal turnover: a 427–452. 1375 1296 Late Miocene record of decoupled land-ocean evolution. Geobios 46, Bernor, R.L., Fahlbusch, V., Andrews, P., De Bruijn, H., Fortelius, M., Rögl, F., 1376 1297 151–157. Steininger, F.F., Werdelin, L., 1996. The evolution of western Eurasian 1377 1298 Alba, D.M., Casanovas-Vilar, I., Robles, J.M., Moyà-Solà, S., 2011. Parada 2. mammal faunas: a chronologic, systematic, biogeographic and pale- 1378 1299 El Aragoniense superior y la transición con el Vallesiense: Can Mata oenvironmental review. In: Bernor, R.L., Fahlbusch, V., Mittmann, 1379 1300 y la exposición paleontológica de els Hostalets de Pierola. Paleontol. H.-W. (Eds.), The Evolution of Western Eurasian Neogene Mammal 1380 1301 Evol. 6, 95–109. Faunas. Columbia University Press, New York, pp. 449–469. 1381 1302 Alba, D.M., Carmona, R., Bertó Mengual, J.V., Casanovas-Vilar, I., Furió, M., Bernor, R.L., Kordos, L., Rook, L., 2005. Multidiscipinary research at Rud- 1382 1303 Garcés, M., Galindo, J., Luján, A.H., 2012a. Intervenció paleontológica abánya. Paleontogr. Ital. 90, 1–313. 1383 1304 a l’Ecoparc de Can Mata (els Hostalets de Pierola, conca del Vallès- Bessedik, M., Cabrera, L., 1985. Le couple récif-mangrove a Sant Pau d’Ordal 1384 1305 Penedès). Trib. Arqueol. 2010–2011, 115–130. (Vallès-Penedès), témoin du maximum transgressif en Mediterranée 1385 1306 Alba, D.M., Moyà-Solà, S., Robles, J.M., Galindo, J., 2012b. Brief communica- nord occidentale (Burdigalien supérieur–Langhien inferieur). Newsl. 1386 1307 tion: the oldest pliopithecid record in the Iberian Peninsula based on Stratigr. 14, 20–35. 1387 1308 new material from the Vallès-Penedès Basin. Am. J. Phys. Anthropol. Bulot, C., Ginsburg, L., 1993. Gisements à mammifères miocènes du Haut- 1388 1309 147, 135–140. Armanag et âge des plus anciens proboscidiens d’Europe occidentale. 1389 1310 Alba, D.M., Fortuny, J., Pérez de los Ríos, M., Zanolli, C., Almécija, S., C. R. Acad. Sci. Paris, Ser. II 316, 1011–1016. 1390 1311 Casanovas-Vilar, I., Robles, J.M., Moyà-Solà, S., 2013. New dental

20 I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx

1391 Cabrera, L., 1981a. Estratigrafía y características sedimentológicas Eronen, J.T., Ataabadi, M.M., Micheels, A., Karme, A., Bernor, R.L., Fortelius, 1470 1392 generales de las formaciones continentales de la cuenca del Vallès- M., 2009. Distribution history and climatic controls of the Late 1471 1393 Penedès (Barcelona, Espana).˜ Estudios Geol. 37, 35–43. Miocene Pikermian chronofauna. Proc. Natl. Acad. Sci. U S A 106, 1472 1394 Cabrera, L., 1981b. Influencia de la tectónica en la sedimentación continen- 11867–11871. 1473 1395 tal de la cuenca del Vallès-Penedès (provincia de Barcelona, Espana)˜ Flynn, L.J., Wessels, W., 2013. Paleobiogeography and South Asian small 1474 1396 durante el Mioceno Inferior. Acta Geol. Hisp. 16, 163–169. mammals: Neogene latitudinal and faunal variation. In: Wang, X., 1475 1397 Cabrera, L., Calvet, F., 1996. Onshore Neogene record in NE Spain: Vallès- Flynn, L.J., Fortelius, M. (Eds.), Fossil Mammals of Asia. Neogene Bios- 1476 1398 Penedès and El Camp half grabens (NW Mediterranean). In: Friend, tratigraphy and Chronology. Columbia University Press, New York, pp. 1477 1399 P.F., Dabrio, C.J. (Eds.), Tertiary Basins of Spain the Stratigraphic Record 445–460. 1478 1400 of Crustal Kinematics. Cambridge University Press, Cambridge, pp. Flynn, L.J., Barry, J.C., Morgan, M.E., Pilbeam, D., Jacobs, L.L., Lindsay, 1479 1401 97–105. E.H., 1995. Neogene Siwalik mammalian lineages: species longevities, 1480 1402 Cabrera, L., Calvet, F., Guimerà, J., Permanyer, A., 1991. El registro sed- rates of change and mode of speciation. Palaeogeogr. Palaeoclimatol. 1481 1403 imentario miocénico en los semigrabens del Vallès-Penedès y de El Palaeoecol. 115, 249–264. 1482 1404 Camp: Organización secuencial y relaciones tectónica sedimentación. Flynn, L.J., Lindsay, E.H., Pilbeam, D., Mahmood Raza, S., Morgan, M.E., 1483 1405 In: Colombo, F. (Ed.), Libro Guía no. 4 del I Congreso Grupo Espanol˜ Barry, J.C., Badgley, C.E., Behrensmeyer, A.K., Cheema, I.U., Rahim Raj- 1484 1406 del Terciario. Vic (132 p). par, A., Opdyke, N.D., 2013. The Siwaliks and Neogene evolutionary 1485 1407 Cabrera, L., Roca, E., Garcés, M., de Porta, J., 2004. Estratigrafía y evolución biology in South Asia. In: Wang, X., Flynn, L.J., Fortelius, M. (Eds.), Fossil 1486 1408 tectonosedimentaria oligocena superior-neógena del sector central Mammals of Asia. Neogene Biostratigraphy and Chronology. Columbia 1487 1409 del margen catalán (Cadena Costero-Catalana). In: Vera, J.A. (Ed.), University Press, New York, pp. 353–372. 1488 1410 Geología de Espana.˜ SGE-IGME, Madrid, pp. 569–573. Foote, M., 2000. Origination and extinction components of taxonomic 1489 1411 Casanovas-Vilar, I., Agustí, J., 2007. Ecogeographical stability and climate diversity: general problems. Paleobiology 26 (4), 74–102. 1490 1412 forcing in the Late Miocene (Vallesian) rodent record of Spain. Palaeo- Fortelius, M., Hokkanen, A., 2001. The trophic context of hominoid occur- 1491 1413 geogr. Palaeoclimatol. Palaeoecol. 248, 169–189. rence in the later Miocene of western Eurasia–a primate-free view. 1492 1414 Casanovas-Vilar, I., Alba, D.M., Moyà-Solà, S., Galindo, J., Cabrera, L., Garcés, In: de Bonis, L., Koufos, G., Andrews, P. (Eds.), Hominoid Evolution and 1493 1415 M., Furió, M., Robles, J.M., Köhler, M., Angelone, C., 2008. Biochronolog- Climatic Change in Europe. Vol. 2: Phylogeny of the Neogene Homi- 1494 1416 ical, taphonomical and paleoenvironmental background of the great noid Primates of Eurasia. Cambridge University Press, Cambridge, pp. 1495 1417 ape Pierolapithecus catalaunicus (Primates, Hominidae). J. Hum. Evol. 19–47. 1496 1418 55, 589–603. Fortelius, M., Werdelin, L., Andrews, P., Bernor, R.L., Gentry, A., Humphrey, 1497 1419 Casanovas-Vilar, I., García-Paredes, I., Alba, D.M., van den Hoek Ostende, L., Mittmann, H.-W., Viratana, S., 1996. Provinciality, diversity, 1498 1420 L.W., Moyà-Solà, S., 2010. The European Far West: Miocene mammal turnover, and paleoecology in land mammal faunas of the Later 1499 1421 isolation, diversity and turnover in the Iberian Peninsula. J. Biogeogr. Miocene of western Eurasia. In: Bernor, R.L., Fahlbusch, V., Mittmann, 1500 1422 37, 1079–1093. H.-W. (Eds.), The Evolution of Western Eurasian Neogene Mammal 1501 1423 Casanovas-Vilar, I., Alba, D.M., Robles, J.M., Moyà-Solà, S., 2011a. Registro Faunas. Columbia University Press, New York, pp. 414–448. 1502 1424 paleontológico continental del Mioceno de la Cuenca del Vallès- Franzen, J.L., Pickford, M., Costeur, L., 2013. Palaeobiodiversity, palaeoe- 1503 1425 Penedès. Paleontol. Evol. 6, 55–80. cology, palaeobiogeography and biochronology of Dorn-Dürkheim 1 1504 1426 Casanovas-Vilar, I., DeMiguel, D., Galindo, J., Robles, J.M., Garcés, M., Cabr- – a summary. Palaeobio. Palaeoenv. 93, 277–284. 1505 1427 era, L., 2011b. The continental Burdigalian (Early Miocene) of the Freudenthal, M., 1967. On the mammalian fauna of the Hipparion-beds 1506 1428 Vallès-Penedès Basin (Catalonia, Spain). In: Pérez-García, A., Gascó, in the Calatayud-Teruel Basin. Part 3: The genera Democricetodon 1507 1429 F., Gasulla, J.M., Escaso, F. (Eds.), Viajando a Mundos Pretéritos. Ayun- and Rotundomys (Rodentia). Proc. Kon. Ned. Akad. Wetensch. B 70, 1508 1430 tamiento de Morella, Morella, pp. 93–100. 298–315. 1509 1431 Casanovas-Vilar, I., Alba, D.M., Moyà-Solà, S., 2011c. Parada 1: Panorámica Freudenthal, M., Mein, P., Martín-Suárez, E., 1998. Revision of the Late 1510 1432 general de la cuenca y yacimiento de els Casots (Subirats): una fauna Miocene and Pliocene Cricetinae (Rodentia, Mammalia) from Spain 1511 1433 de vertebrados del Mioceno Inferior. Paleontol. Evol. 6, 81–88. and France. Treb. Mus. Geol. Barcelona 7, 11–93. 1512 1434 Casanovas-Vilar, I., Alba, D.M., Garcés, M., Robles, J.M., Moyà-Solà, S., Furió, M., Casanovas-Vilar, I., Moyà-Solà, S., Köhler, M., Galindo, J., Alba, 1513 1435 2011d. Updated chronology of the Miocene hominoid radiation in D.M., 2011. Insectivores (Eulipotyphla; Mammalia) from the Middle 1514 1436 Western Eurasia. Proc. Natl. Acad. Sci. U S A 108, 5554–5559. Miocene of Barranc de Can Vila 1 (Vallès-Penedès Basin, Catalonia, 1515 1437 Casanovas-Vilar, I., Van den Hoek Ostende, L.W., Furió, M., Madern, P.A., Spain). Geobios 44, 199–213. 1516 1438 2014. The range and extent of the Vallesian Crisis (Late Miocene): new Furió, M., Prieto, J., van den Hoek Ostende, L.W., 2015. Three million 1517 1439 prospects based on the micromammal record from the Vallès-Penedès years of “Terror-Shrew” (Dinosorex, Eulipotyphla, Mammalia) in the 1518 1440 basin (Catalonia, Spain). J. Iber. Geol. 40, 29–48. Miocene of the Vallès-Penedès Basin (Barcelona, Spain). C. R. Palevol. 1519 1441 Crusafont Pairó, M., 1950. La cuestión del llamado Meótico espanol.˜ Arra- 14, 111–124. 1520 1442 hona 1950, 41–48. Gallart, F., 1981. Neógeno superior y Cuaternario del Penedès. Acta Geol. 1521 1443 Crusafont Pairó, M., Truyols Santonja, J., 1954. Sinopsis estratigráfico- Hisp. 16, 151–157. 1522 1444 paleontológica del Vallès-Penedès. Arrahona 1954, 1–15. Garcés, M., Agustí, J., Cabrera, L., Parés, J.M., 1996. Magnetostratigraphy 1523 1445 Crusafont Pairó, M., Truyols Santonja, J., 1960. Sobre la caracterización del of the Vallesian (late Miocene) in the Vallès-Penedès Basin (northeast 1524 1446 Vallesienese. Notas y Comm. Inst. Geol. Min. Espana˜ 60, 109–126. Spain). Earth Planet. Sci. Lett. 142, 381–396. 1525 1447 Crusafont Pairó, M., de Villalta, J.F., 1951. Los nuevos mamíferos del Garcés, M., Cabrera, L., Agustí, J., Parés, J.M., 1997. Old World first 1526 1448 Neogeno de Espana.˜ Notas y Com. Inst. Geol. Min. Esp. 22, 127–151. appearance datum of “Hipparion” horses: late Miocene large mammal 1527 1449 Crusafont, M., de Villalta, J.F., Truyols, J., 1955. El Burdigaliense continental dispersals and global events. Geology 25, 19–22. 1528 1450 de la Cuenca del Vallés-Penedés. Mem. Com. Inst. Geol. 12, 1–272. Garcés, M., Krijgsman, W., Peláez-Campomanes, P., Álvarez Sierra, M.A., 1529 1451 Daams, R., Freudenthal, M., 1974. Early Miocene Cricetidae (Rodentia, Daams, R., 2003. Hipparion dispersal in Europe: magnetostratigraphic 1530 1452 Mammalia) from Bunol˜ (Prov. Valencia, Spain). Scripta Geol. 24, 1–19. constraints from the Daroca area (Spain). Col. Paleontol. 1, 171–178. 1531 1453 Daams, R., Freudenthal, M., 1988. Synopsis of the Dutch-Spanish collbra- de Gibert, J.M., Casanovas-Vilar, I., 2011. Contexto geológico del Mioceno 1532 1454 tion program in the Aragonian type area. Scripta Geol. 1, 3–18. de la Cuenca del Vallès-Penedès. Paleontol. Evol. 6, 39–45. 1533 1455 Daams, R., Freudenthal, M., 1989. The Ramblian and the Aragonian: limits, Golpe-Posse, J.M., 1974. Faunas de yacimientos con suiformes del Terciario 1534 1456 subdivision, geographical and temporal extension. In: Lindsay, E.H., espanol.˜ Paleontol. Evol. 2, 1–197. 1535 1457 Fahlbusch, V., Mein, P. (Eds.), European Neogene Mammal Chronology. Hilgen, F.J., Lourens, L.J., Van Dam, J.A., 2012. The Neogene period. In: Grad- 1536 1458 Plenum Press, New York, pp. 51–59. stein, F.M., Ogg, J.G., Schmitz, M., Ogg, G. (Eds.), The Geologic Time 1537 1459 Daams, R., Freudenthal, M., Van der Meulen, A.J., 1988. Ecostratigraphy Scale 2012. Elsevier, Amsterdam, pp. 923–978. 1538 1460 of micromammal faunas from the Neogene of the Calatayud-Teruel Institut Cartogràfic i Geològic de Catalunya, 2015. www.icc.cat (accessed 1539 1461 Basin. Scripta Geol. 1, 287–302. on January 2015). 1540 1462 Daams, R., Meulen, A.J., Van der Álvarez Sierra, M.A., Peláez-Campomanes, Jacobs, L.L., 1977. A new genus of murid rodent from the Miocene of 1541 1463 P., Pedro Calvo, J., Alonso Zarza, M.A., Krijgsman, W., 1999. Stratigra- Pakistan and comments on the origin of the Muridae. Paleobios 25, 1542 1464 phy and sedimentology of the Aragonian (Early to Middle Miocene) in 1–11. 1543 1465 its type area (North-Central Spain). Newsl. Stratigr. 37, 103–139. Johnson, N.M., Sheikh, K.A., Dawson-Saunders, E., McRae, L.E., 1988. The 1544 1466 DeMiguel, D., Azanza, B., Morales, J., 2011. Paleoenvironments and paleo- use of magnetic-reversal time lines in stratigraphic analysis: a case 1545 1467 climate of the Middle Miocene of central Spain: A reconstruction from study in measuring variability in sedimentation rate. In: Kleinspehn, 1546 1468 dental wear of ruminants. Palaeogeogr. Palaeoclimatol. Palaeoecol. K.L., Paola, C. (Eds.), New Perspectives in Basin Analysis. Springer, New 1547 1469 302, 452–463. York, pp. 189–200. 1548

I. Casanovas-Vilar et al. / C. R. Palevol xxx (2015) xxx–xxx 21

1549 Kappelman, J., Duncam, A., Feseha, M., Lunkka, J.-P., Ekart, D., McDowell, Ruiz-Sánchez, F., de Santisteban Bové, C., Lacomba Andueza, J.I., 2003. 1626 1550 F., Ryan, T.M., Swisher III., C.C., 2003. Chronology. In: Fortelius, M., Nuevas faunas de roedores fósiles (Mammalia, Rodentia) de edad 1627 1551 Kappelman, J., Sen, S., Bernor, R.L. (Eds.), Geology and Paleontology Aragoniense inferior y medio en la serie del Barranco de Morteral 1628 1552 of the Miocene Sinap Formation, Turkey. Columbia University Press, (cuenca del Río Magro, prov. de Valencia, Espana).˜ Col. Paleontol. 1, 1629 1553 New York, pp. 41–66. 579–594. 1630 1554 Krijgsman, W., Garcés, M., Langereis, C.G., Daams, R., Van Dam, J., Van der Sadler, P.M., 1981. Sediment accumulation rates and the completeness of 1631 1555 Meulen, A.J., Agustí, J., Cabrera, L., 1996. A new chronology for the stratigraphic sections. J. Geol. 89, 569–584. 1632 1556 Middle to Late Miocene continental record in Spain. Earth Planet. Sci. Sanz de Siria Catalán, A., 1993. Datos sobre la paleoclimatología y paleoe- 1633 1557 Lett. 142, 367–380. cología del Neógeno del Vallès-Penedès según las macroﬂoras halladas 1634 1558 Larrasoana,˜ J.C., Murelaga, X., Garcés, M., 2006. Magnetobiochronology of en la cuenca y zonas próximas. Paleontol. Evol. 26–27, 281–289. 1635 1559 Lower Miocene (Ramblian) continental sediments from the Tudela Sanz de Siria Catalán, A., 1994. La evolución de las paleoﬂoras en las cuen- 1636 1560 formation (western Ebro basin, Spain). Earth Planet. Sci. Lett. 243, cas cenozoicas catalanas. Acta Geol. Hisp. 29, 169–189. 1637 1561 409–423. Sanz de Siria Catalán, A., 2001. Flora y vegetación del Mioceno medio de 1638 1562 Maridet, O., Wu, W.-Y., Ye, J., Bi, S.-D., Ni, X.-J., Meng, J., 2011. Earli- la depresión del Vallès-Penedès. Paleontol. Evol. 32–33, 79–92. 1639 1563 est occurrence of Democricetodon in China, in the Early Miocene of Solounias, N., Plavcan, J.M., Quade, J., Witmer, L., 1999. The paleoecology 1640 1564 the Junggar Basin (Xinjiang, China) and comparison with the genus of the Pikermian Biome and the savanna myth. In: Agustí, J., Rook, 1641 1565 Spanocricetodon. Vert. PalAsiat. 49, 393–405. L., Andrews, P. (Eds.), Hominoid Evolution and Climatic Change in 1642 1566 Marmi, J.M., Casanovas-Vilar, I., Robles, J.M., Moyà-Solà, S., Alba, D.M., Europe. Vol. 1: The Evolution of Neogene Terrestrial Ecosystems in 1643 1567 2012. The paleoenvironment of Hispanopithecus laietanus as revealed Europe. Cambridge University Press, Cambridge, pp. 436–453. 1644 1568 by paleobotanical evidence from the Late Miocene of Can Llobateres Steininger, F.F., 1999. Chronostratigraphy, geochronology and biochronol- 1645 1569 1 (Catalonia, Spain). J. Hum. Evol. 62, 412–423. ogy of the Miocene “European Land Mammal Mega-Zones” (ELMMZ) 1646 1570 Martinell, J., 1988. An overview of the marine Pliocene of NE Spain. Geol. and the Miocene “Mammal-Zones (MN-Zones)”. In: Rössner, G.E., 1647 1571 Mediterr. 15, 227–233. Heissig, K. (Eds.), The Miocene Land Mammals of Europe. Verlag Dr. 1648 1572 Mein, P., 1975a. Résultats du groupe de travail des vertébrés. In: Sénes, Riedrich Pfeil, Munich, pp. 9–24. 1649 1573 J. (Ed.), Report on the Activity of R. C. M. N. S. Working Groups–6. Van Dam, J.A., 1997. The small mammals from the Upper Miocene of 1650 1574 Congress of the Regional Committee of Mediterranian Nerogene the Teruel-Alfambra region (Spain): paleobiology and paleoclimatic 1651 1575 Stratigraphy. SAV, Bratislava, pp. 78–81. reconstructions. Geol. Ultraiectina 156, 4–204. 1652 1576 Mein, P., 1975b. Une forme de transition entre deux familles de rongeurs. Van Dam, J.A., 2003. European Neogene mammal chronology: past, 1653 1577 Coll. Int. C. N. R. S 218, 759–763. present and future. DEINSEA 10, 85–95. 1654 1578 Mein, P., 1999. European Miocene Mammal Biochronology. In: Rössner, Van Dam, J.A., Alcalá, L., Alonso-Zarza, A., Calvo, J.P., Garcés, M., Krijgsman, 1655 1579 G.E., Heissig, K. (Eds.), The Miocene Land Mammals of Europe. Verlag W., 2001. The upper Miocene mammal record from the Teruel- 1656 1580 Dr. Riedrich Pfeil, Munich, pp. 25–38. Alfambra region (Spain). The MN system and continental stage/age 1657 1581 Morales, J., Nieto, M., Köhler, M., Moyà-Solà, S., 1999. Large mammals from concepts discussed. J. Vert. Paleontol. 21, 367–385. 1658 1582 the Vallesian of Spain. In: Agustí, J., Rook, L., Andrews, P. (Eds.), Homi- Van Dam, J.A., Abdul Aziz, H., Álvarez-Sierra, M.A., Hilgen, F.J., Van den 1659 1583 noid Evolution and Climatic Change in Europe. Vol. 1: The Evolution Hoek Ostende, L.W., Lourens, L.J., Mein, P., Van der Meulen, A.J., Pélaez- 1660 1584 of Neogene Terrestrial Ecosystems in Europe. Cambridge University Campomanes, P., 2006. Long-period astronomical forcing of mammal 1661 1585 Press, Cambridge, pp. 113–126. turnover. Nature 443, 687–691. 1662 1586 Mossbrugger, V., Utescher, T., Dilcher, D.L., 2005. Cenozoic climatic evolu- Van Dam, J.A., Krijgsman, W., Abels, H.A., Álvarez-Sierra, M.A., García- 1663 1587 tion of central Europe. Proc. Natl. Acad. Sci. U S A 102, 14964–14969. Paredes, I., López-Guerrero, P., Pélaez-Campomanes, P., Ventra, D., 1664 1588 Moyà-Solà, S., Agustí, J., 1987. The Vallesian in the type area (Vallès- 2014. Updated chronology for Middle to Late Miocene mammal sites 1665 1589 Penedès, Barcelona, Spain). Ann. Inst. Geol. Publ. Hung. 70, 93–99. of the Daroca area (Calatayud-Montalbán Basin, Spain). Geobios 47, 1666 1590 Moyà-Solà, S., Agustí, J., 1989. Bioevents and mammal successions in the 325–334. 1667 1591 Spanish Miocene. In: Lindsay, E.H., Fahlbusch, V., Mein, P. (Eds.), Euro- Van den Hoek Ostende, L.W., Furió, M., 2005. Spain. Scripta Geol. 5, 1668 1592 pean Neogene Mammal Chronology. Plenum Press, New York, pp. 175–203. 1669 1593 357–373. Van der Meulen, A.J., De Bruijn, H., 1982. The mammals from the Lower 1670 1594 Moyà-Solà, S., Köhler, M., Alba, D.M., 2001. Egarapithecus narcisoi, a new Miocene of Aliveri (Island of Evia, Greece). Part 2: The Gliridae. Proc. 1671 1595 genus of Pliopithecidae (Primates, Catarrhini) from the Late Miocene Kon. Ned. Akad. Wetensch. B 85, 485–524. 1672 1596 of Spain. Am. J. Phys. Anthropol. 114, 312–324. Van der Meulen, A.J., Pélaez-Campomanes, P., Levin, S.A., 2005. Age struc- 1673 1597 Moyà-Solà, S., Köhler, M., Alba, D.M., Casanovas-Vilar, I., Galindo, J., Rob- ture, residents, and transients of Miocene rodent communities. Am. 1674 1598 les, J.M., Cabrera, L., Garcés, M., Almécija, S., Beamud, E., 2009. First Nat. 165, E108–E125. 1675 1599 partial face and upper dentition of the Middle Miocene hominoid Dry- Van der Meulen, A.J., García-Paredes, I., Álvarez-Sierra, M.A., Van den 1676 1600 opithecus fontani from Abocador de Can Mata (Vallès-Penedès Basin, Hoek Ostende, L.W., Hordijk, K., Oliver, A., López-Guerrero, P., 1677 1601 Catalonia, NE Spain). Am. J. Phys. Anthropol. 139, 126–145. Hernández-Ballarín, V., Pélaez-Campomanes, P., 2011. Biostratigra- 1678 1602 Ogg, J.G., 2012. Geomagnetic Polarity Time Scale. In: Gradstein, F.M., Ogg, phy or biochronology? Lessons from the Early and Middle Miocene 1679 1603 J.G., Schmitz, M., Ogg, G. (Eds.), The Geologic Time Scale 2012. Elsevier, small mammal events in Europe. Geobios 44, 309–321. 1680 1604 Amsterdam, pp. 85–113. Van der Meulen, A.J., García-Paredes, I., Álvarez-Sierra, M.A., Van den Hoek 1681 1605 Opdyke, N.D., Channel, J.E.T., 1996. Magnetic Stratigraphy. Academic Press, Ostende, L.W., Hordijk, K., Oliver, A., Pélaez-Campomanes, P., 2012. 1682 1606 San Diego (346 p). Updated Aragonian biostratigraphy: small mammal distribution and 1683 1607 Pálfy, J., Mundil, R., Renne, P.R., Bernor, R.L., Kordos, L., Gasparik, M., 2007. its implications for the Miocene European chronology. Geol. Acta 10, 1684 1608 U-Pb and 40Ar/39Ar dating of the Miocene fossil track site at Ipoly- 159–179. 1685 1609 tarnóc (Hungary) and its implications. Earth Planet. Sci. Lett. 258, Van de Weerd, A., 1976. Rodent faunas of the Mio-Pliocene continental 1686 1610 160–174. sediments of the Teruelo-Alfambra region, Spain. Utrecht Micropale- 1687 1611 Robles, J.M., Alba, D.M., Casanovas-Vilar, I., Galindo, J., Cabrera, L., Car- ontol. Bull. 2, 1–185. 1688 1612 mona, R., Moyà-Solà, S., 2011. On the age of the paleontological site de Villalta Comella, J.F., Crusafont Pairó, M., 1941a. Los vertebrados del 1689 1613 of Can Missert (Terrassa, Vallès-Penedès Basin, NE Iberian Penin- Mioceno continental del Vallès-Penedès (Provincia de Barcelona). I: 1690 1614 sula). In: Pérez-García, A., Gascó, F., Gasulla, J.M., Escaso, F. (Eds.), Insectívoros. II: Carnívoros. Bol. Inst. Geol. Min. Esp. 56, 145–336. 1691 1615 Viajando a Mundos Pretéritos. Ayuntamiento de Morella, Morella, de Villalta Comella, J.F., Crusafont Pairó, M., 1941b. Los vertebrados fósiles 1692 1616 pp. 339–346. del Mioceno del Vallès-Penedès (Provincia de Barcelona). Publica- 1693 1617 Roca, E., Guimerà, J., 1992. The Neogene structure of the eastern Iberian ciones del Museo de Sabadell, Sabadell (16 p). 1694 1618 margin: structural constraints on the crustal evolution of the Valencia Wessels, W., 2009. Miocene rodent evolution and migration: Muroidea 1695 1619 Trough (western Mediterranean). Tectonophysics 203, 203–218. from Pakistan, Turkey and Northern Africa. Geol. Ultraiectina 307, 1696 1620 Roca, E., Sans, M., Cabrera, L., Marzo, M., 1999. Oligocene to Mid- 1–290. 1697 1621 dle Miocene evolution of the central Catalan margin (northwestern Woodburne, M.O., 2009. The early Vallesian vertebrates from Atzelsdorf 1698 1622 Mediterranean). Tectonophysics 315, 209–233. (Late Miocene, Austria). 9. Hippotherium (Mammalia, Equidae). Ann. 1699 1623 Rotgers, C., Alba, D.M., 2011. The genus Anchitherium (Equidae: Naturhistorisch. Mus. Wien 111A, 585–604. 1700 1624 Anchitheriinae) in the Vallès-Penedès Basin (Catalonia, Spain). In: Zachos, J., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, 1701 1625 Pérez-García, A., Gascó, F., Gasulla, J.M., Escaso, F. (Eds.), Viajando a rhythms and aberrations in global climate 65 Ma to present. Science 1702 Mundos Pretéritos. Ayuntamiento de Morella, Morella, pp. 347–354. 292, 686–693. 1703