This is the accepted version of the article: Březina, Jakub; Alba, David M.; Ivanov, Martin; [et al.]. «A Middle Miocene vertebrate assemblage from the Czech part of the Vienna Basin : implica- tions for the paleoenvironments of the Central Paratethys». Palaeogeography, Palaeoclimatology, Palaeoecology, Vol. 575 (August 2021), art. 110473. DOI /10.1016/j.palaeo.2021.110473 This version is available at https://ddd.uab.cat/record/241122 under the terms of the license 1 A Middle Miocene vertebrate assemblage from the Vienna Basin of the 2 southern Czech Republic: implications for the paleoenvironments of the 3 Central Paratethys 4 5 Jakub Březinaa,b,*, David M. Albac, Martin Ivanovb, Martin Hanáčekd, Àngel H. Lujánb,c 6 a 7 Department of Geology and Paleontology, Moravian Museum, Zelný trh 6, 659 37 Brno, 8 Czech Republic b 9 Department of Geological Sciences, Faculty of Science, Masaryk University, Kotlářská 2, 10 611 37 Brno, Czech Republic c 11 Institut Català de Paleontologia Miquel Crusafont, Universitat Autònoma de Barcelona, 12 Edifici ICTA-ICP, c/ Columnes s/n, Campus de la Universitat Autònoma de Barcelona, E- 13 08193 Cerdanyola del Vallès, Barcelona, Spain d 14 Department of Geography, Faculty of Science, Masaryk University, Kotlářská 2, 611 37 15 Brno, Czech Republic 16 * Corresponding author 17 18 E-mail addressess: jbrezina@mzm.cz (J. Březina), david.alba@icp.cat (D.M. Alba), 19 mivanov@sci.muni.cz (M. Ivanov), HanacekM@seznam.cz (M. Hanáček), 20 angel.lujan@icp.cat (À.H. Luján) 21 22 1 23 Abstract 24 Czujan’s sandpit is an abandoned quarry in the Vienna Basin (Mikulov, Czech 25 Republic) that has yielded an important middle Miocene vertebrate assemblage. Here we re- 26 describe the site from the perspective of sedimentology, taphonomy, and paleoenvironments, 27 and further review the biochronology of the fauna to clarify the age. The updated faunal list 28 includes two testudines (one trionychid and one medium-sized testudinid), and 12 species of 29 terrestrial mammals (three proboscideans, four perissodactyls, four artiodactyls, and one 30 carnivoran), consistent with an early Astaracian (MN6) age. The position of the 31 Wielician/Kosovian boundary just below the floor of Czujan’s sandpit, and our new 32 biostratigraphic data, further allow us to constrain the fossil assemblage to the latest MN6 33 (late Badenian, ⁓13.6 Ma) and resolves a longstanding controversy about the age of the site. 34 The site exposes a coarsening-upward succession deposited in a braid delta environment, and 35 comprises three facies association: from bottom to top, pelagic sediments (FA1); prodelta and 36 delta slope sediments (FA2); and distributory channel infills of the delta front and delta plain 37 (FA3), the latter containing all the studied terrestrial vertebrates. We propose two taphonomic 38 explanations for the genesis of the vertebrate assemblage: (1) a time-averaged assemblage 39 generated by riverine transport, or (2) a transported assemblage from a mass death site(s), 40 with mass death episode(s) caused by seasonal droughts in the river catchment. Our new 41 findings allow the more precise reconstruction of late Badenian terrestrial paleoenvironments 42 in the northwest area of the Vienna Basin and adjacent Carpathian Foredeep Basin. This 43 region comprised a mosaic of continental habitats dominated by woodlands but also including 44 forest patches and more open environments. 45 46 Keywords 47 Badenian; Mikulov; Mammalia; Testudines; Taphonomy; Paleoecology 2 48 49 1. Introduction 50 The intramontane Vienna Basin is among the best explored Paratethyan basins, where 51 decades of research of both marine and brackish sedimentary deposits have greatly 52 contributed to the understanding of paleoenvironmental changes during the middle to late 53 Miocene (e.g., Jiříček, 2002; Kováč et al., 2007, 2018; Harzhauser et al., 2011, 2018). 54 Paleoenvironmental reconstructions of the middle Miocene environment are largely based on 55 the marine fauna (e.g., Harzhauser et al., 2011) or palynoflora (e.g., Kvaček et al., 2006; 56 Kováčová et al., 2011; Kovar-Eder and Teodoridis, 2018). Middle Miocene sites with 57 terrestrial vertebrates are rather scarce with only two well-documented localities, comprising 58 Klein-hadersdorf, Austria (e.g., Thenius, 1948, 1951; Böhme et al., 2012) situated close to the 59 studied site in western part of the Vienna Basin; and Devínska Nová Ves, Slovakia (e.g., 60 Zapfe, 1949, 1954, 1979, 1993; Estes, 1969; Sabol and Holec, 2002; Sabol and Kováč, 2006) 61 situated at the southeastern margin of the Vienna Basin. A detailed paleoenvironmental 62 analysis based on the complete spectrum of the terrestrial taphocoenosis has been conducted 63 only for the Devínska Nová Ves-“Bonanza” site (Sabol and Kováč, 2006). 64 Here we revise the evidence from another, less known, vertebrate site called Czujan’s 65 sandpit (sand quarry), which is situated close to the Carpathian Foredeep Basin at the 66 northwestern margin of the Vienna Basin (Fig. 1). This site has yielded a high concentration 67 of fossil macrovertebrates. However, this relatively diverse assemblage has remained largely 68 unpublished, even though it constitutes one of the most important middle Miocene mammal 69 assemblages from the Central Paratethys. The importance of this site increased in the 1950s, 70 when Thenius (1951) established it as a type locality of the bovid Tethytragus stehlini. Later, 71 the description of the proboscidean fauna (Holec, 1985; Seitl, 1985) highlighted this locality 72 as one of the most important occurrences of the relatively rare proboscidean Zygolophodon 3 73 turicensis. Despite the significance of the studied site, the fossil fauna, its taphonomy and 74 sedimentology has never been studied comprehensively. Accordingly, the research carried out 75 since the 1930s did not provide a clear agreement on the stratigraphical position of this site 76 (e.g., Seitl, 1985; Čtyroký et al., 1990; Stráník et al., 1999). Moreover, Late Miocene 77 mammals are known from other abandoned sandpits (Fig. 1B) in the southern area of Mikulov 78 as well. Although multiple sites near Mikulov have been grouped as a single locality, Mikulov 79 (= Nikolsburg), and considered contemporaneous (e.g., Abel, 1910; Rzehak, 1922; Stejskal, 80 1934; Thenius, 1959; Fejfar, 1990; Böhme et al., 2012), Middle and Late Miocene vertebrates 81 have been found in various sites of the Mikulov area (Čtyroký, 1989; Čtyroký et al., 1990; 82 Březina, 2019). Therefore, studied sites in the Mikulov area must be always clearly specified. 83 Here we provide the first taxonomic review of the fossil vertebrates from the former 84 Czujan’sandpit. In addition, the combination of sedimentological, taphonomic and 85 taxonomical characteristics allows us to (1) determine the age of the site; and (2) provide a 86 reconstruction of the late Badenian terrestrial paleoenvironment of the northwestern part of 87 the Vienna Basin and adjacent southeastern part of the Carpathian Foredeep Basin, in areas 88 where the paleobotanical fossil record is rather poor. 89 90 2. Czujan’s sandpit and overview of research on Neogene vertebrates from the Mikulov 91 area 92 The former Czujan’s sandpit (also called Furstenälle, Fürstenwegfeld, Teichäcker or 93 Na Rybníkách) was a psephitic and psammitic quarry located 2 km east of Mikulov (Fig. 1A– 94 B; GPS: 48°47'49" N, 16°40'18" E) in the South Moravian Region, Czech Republic. The 95 mining activity began in the 1930s, reaching a maximum extension of 250 × 120 m and 7–10 96 m in depth. Exploitation of sands and gravels progressed from south to northwest and the 97 abandoned southern parts of the sandpit were gradually filled with garbage (Březina, 2019). 4 98 During the 1980s, Czujan’s sandpit was filled to the top and leveled completely, being 99 currently part of a vineyard. 100 Czujan’s sandpit stands out as the oldest Miocene vertebrate site of the Mikulov area, 101 which is located close the Carpathian Foredeep Basin. Indeed, vertebrate fossil remains have 102 been known from sandpits in the Mikulov area since the 19th century (e.g., Hörnes, 1848; 103 Alth, 1850). However, their contribution to a broader understanding of the European middle 104 Miocene paleoenvironments and faunas has been hampered by different conceptions about the 105 age of the various sites. Abel (1910) advocated an early Pliocene age, given the faunal 106 similarities with Eppelsheim (currently late Miocene, Pannonian), including the presence of 107 Deinotherium giganteum, Chalicotherium aff. goldfussi, Hipparion gracile, and Chalicomys 108 jaegeri. Stejskal (1934) agreed with the early Pliocene age suggested by Abel (1910), whereas 109 Rzehak (1922) dated it to the late Pliocene based on the absence of Pannonian mollusks 110 (Congeria and melanopsids). Jüttner (1938) first noted that the Mikulov area included several 111 Miocene vertebrate sites of different ages. In particular, Jüttner (1939a) interpreted Czujan’s 112 sandpit as being Sarmatian in age, given petrographical similarities with the sands from 113 Chrastiny Hill, near Valtice, where Sarmatian mollusks are present. Such interpretation was 114 followed by many subsequent authors (e.g., Musil, 1956; Thenius, 1959; Seitl, 1985). 115 However, Buday et al. (1964) considered Czujan’s sandpit as the highest strata of the late 116 Badenian (Kosovian substage) based on the foraminifers and mollusks. A Kosovian age was 117 later confirmed by comparison with nearby (800 m away) Nové Mlýny-2 borehole (Čtyroký, 118 1989; Fejfar, 1990; Čtyroký et al., 1990). Alternatively, some authors advocated that Czujan’s 119 sandpit is “middle” Badenian (Wielician substage) in age, but such proposals were not 120 substantitated (Kuklová, 1978; Stráník et al., 1999; Jiříček, 2002), and changes in the 121 Badenian subdivision during time (see Kováč et al., 2018) make this assertion doubtful as 122 well. 5 123 A list of fossil mammals from Czujan’s sandpit was first reported by Thenius (1951), 124 who erected the nominal species Gazella stehlini on the basis of a set of horn cores from 125 Czujan’s sandpit (type locality) and Klein-hadersdorf (Austria), some of which are currently 126 housed at the Paleontological Institute of the University of Vienna (Weinfurter collection). In 127 addition, Thenius (1951) provided a faunal list from the site, including chelonians, 128 proboscideans, rhinocerotids, cervids, and chalicotheriids. Later on, Thenius (1979) erected 129 the genus Caprotragoides to accommodate ?Pseudotragus potwaricus (middle Miocene of 130 Fort Ternan, Kenya) as type species, as well as “Gazella” stehlini from Czujan’s sandpit. 131 Subsequently, Azanza and Morales (1994) tentatively included Caprotragoides stehlini in 132 Tethytragus as ?Tethytragus stehlini. Some proboscidean molars from Czujan’s sandpit have 133 also been described (Holec, 1985; Seitl, 1985) and ascribed to Zygolophodon turicensis and 134 Gomphotherium angustidens. 135 136 3. Geological background 137 Czujan’s sandpit belongs to the Pannonian Basin System and is located on the 138 northwest margin of the Vienna Basin (Fig. 1A). In the Mikulov area, Neogene sedimentation 139 began by the deposition of marine lower Badenian clays (Lanžhot Beds) on the Mesozoic and 140 Paleogene flysch basement (Ždánice Unit). At that time, the Vienna Basin was connected with 141 the Carpathian Foredeep Basin by a narrow marine channel known as the “Mikulov Gate” 142 (Brzobohatý and Stráník, 2012). From a tectonic viewpoint, the Vienna Basin was 143 transformed into a pull-apart basin at the beginning of the “middle” Badenian (sensu 144 Harzhauser et al., 2018). The rivers flowed into the basin from the west through the Mikulov 145 area and apparently eroded the Lanžhot Beds (Stráník et al., 1999; Jiříček, 2002). 146 During the “middle” Badenian, the Sedlec Gravels were deposited discordantly on the 147 Ždánice Unit in the Mikulov area (Čtyroký et al., 1990). The overlying “middle” to “upper” 6 148 Badenian (sensu Harzhauser et al., 2018) marine deposits belong to the Hrušky Formation, 149 which has yielded a rich marine fauna (e.g., Kienberg locality; Brzobohatý et al., 2007). The 150 uppermost part of the Hrušky Formation terminates with alternating continental sands and 151 gravels, which contain the terrestrial vertebrate remains uncovered in Czujan’s sandpit 152 (Čtyroký, 1989; Čtyroký et al., 1990). 153 In the Mikulov area, no Sarmatian deposits are known. The closest occurrences of 154 marine Sarmatian deposits are found at Chrastiny Hill near Valtice (Stráník et al., 1999). 155 Therefore, uppermost parts of the Hrušky Formation and even in the area of Mikulov were 156 eroded during the Sarmatian to the early Pannonian. Then, sands and silts (Bzenec Formation, 157 Pannonian; Fig. 1B) were deposited overlying the Badenian clays of the Vienna Basin and the 158 flysch basement (Ždánice Unit). Miocene sedimentation in the Vienna Basin terminates with 159 the uppermost Pannonian and Pontian sediments, mainly composed of gravels and cross- 160 bedded fluviatile sands (Valtice Beds; Čtyroký, 1989, 1999; Čtyroký et al., 1990). Generally, 161 Miocene sequences are covered by Quaternary deposits, such as fluvial and eolian sediments, 162 as well as paleosols (Čtyroký et al., 1990; Stráník et al., 1999). 163 The Czujan’s sandpit deposits are constituted by fine- to coarse-grained sands 164 displaying a trough cross-stratification (Jüttner, 1939a). They are light gray to yellow in color 165 and mainly composed of quartz, quartzite, feldspar, and intercalated eolian quartz and diverse 166 pebbles (Patočková, 1966). Cerha (1987) noted that sands contain an admixture of gravels and 167 clays up to 50% (Fig. 3). According to Kuklová (1970), deposits overlying the Hrušky 168 Formation (i.e., Bzenec Formation and Valtice Beds, respectively, interpreted as Pannonian 169 and Pontian) have never been recorded from the borehole samples in the surroundings of 170 Czujan’s sandpit (see also Čtyroký, 1989; Březina, 2019). 171 172 4. Materials and methods 7 173 174 4.1 Studied material 175 The fossil vertebrate material from Czujan’s sandpit consists of 464 identified remains 176 collected between 1930 and 1970. They include teeth, skulls, postcranial bones, and turtle 177 shell remains (see complete list in the Supplementary Table 1). The studied material is 178 currently housed in the following institutions: PIUW, Paleontological Institut University 179 Vienna (Weinfurter collection), Austria; MZM, Moravian Museum, Brno, Czech Republic; 180 RMM, Regional Museum in Mikulov, Mikulov, Czech Republic; ÚGV, Department of 181 Geological Sciences, Faculty of Sciences, Masaryk University, Brno, Czech Republic. The 182 material housed in MZM includes the research collection of Rudolf Musil from the 1950s and 183 the remains gathered by private collectors between 1956–1959 (V. Solařík) and in 1965 (J. 184 Maňák). The material housed in MZM was prepared for this study by one of the autors (J.B). 185 However, a detailed description of the available material is beyond the scope of this paper, 186 which focuses instead on the most informative specimens from taxonomic, biostratigraphic, 187 and/or paleoecological viewpoints. 188 189 4.2 Sedimentology, taphonomy and biostratigraphy 190 Sedimentological and taphonomic analyses of Czujan’s sandpit are restricted because 191 the site is no longer accessible and also field documentation is lacking. Sedimentological 192 interpretations were based on Jüttner’s (1938, 1939a, 1940) lithological descriptions; 193 associated information preserved on labels; unpublished reports from the Czech Geological 194 Survey (Geofond, Prague); and old photos from the photoarchive of the Department of 195 Geology and Paleontology, MZM. As for the micropaleontological analysis, some samples of 196 clay and silt preserved in bone and tusk cavities were washed using a sieve with a mesh 197 diameter of 0.062 mm, and inspected using a stereomicroscope NIKON SMZ 1. 8 198 Given that fossil remains were recovered from the surface or the profiles of former 199 Czujan’s sandpit without systematic excavation techniques or any documentation of their 200 provenance or spatial distribution, taphonomical analyses are restricted to the study of the 201 assemblage and bone modification data (see Eberth et al., 2007). A bias toward the collection 202 of the most complete fossils during fieldwork is unlikely, given the high number of rounded, 203 fragmentary and otherwise poorly preserved specimens available among the studied material 204 from MZM. The taphonomic analysis is focused on the degree of abrasion, corrosion, and 205 weathering of the bones. We mainly considered the completeness of the bones and their 206 degree of anatomical articulation with other elements to evaluate their preservation state. The 207 taphonomical terminology and practical approach follows Behrensmeyer (1991). We 208 quantified the minimum number of individuals (MNI) and the minimum number of elements 209 (MNE). Other aspects that were taken into account, include differences in tooth wear, 210 epiphyseal fusion, and morphology of fossil bones. 211 The European Neogene Land Mammal Units (Mein, 1975, 1990, 1999; de Bruijn et 212 al., 1992) do not display constustent boundaries throughout the Europe (e.g. Hilgen et al., 213 2012; Koufos, 2016). Mammal zonation applied for the Central European area was modified 214 according to Steininger (1999), Becker (2003) and Hilgen et al. (2012). Age limits for well 215 dated Central European Miocene localities with their fossil record follows works listed in the 216 Supplementary Information. 217 218 5. Results 219 5.1 Sedimentology 220 The reconstruction of a composite profile was made using data obtained from the 221 individual boreholes in close proximity of the Czujan’s sandpit (Fig. 2), which allow us to 222 define three facies associations (FAs; Fig. 3). The lower part of the composite profile (~80 m 9 223 thick) corresponds to FA1 (Wielician to Kosovian), constituted by clays with occasional sand 224 layers. FA2 (Kosovian) reaches a thickness of ~20 m and overlies FA1. Sporadic layers of 225 gravel appear at the base of the FA2, whereas its upper part consists of interbedded sand and 226 clay layers. The upper part of the profile corresponds to FA3 (Kosovian), which almost 227 reaches 20 m in thickness. It is characterized by the highest portion of coarse-grained deposits 228 (sands and gravels) over the clays. This facies association (FA1–FA3) represents a 229 coarsening-up sequence belonging to the Hrušky Formation. Czujan’s sandpit was opened in 230 the uppermost part of the FA3, which contains the coarsest deposits. Both FA2+FA3 231 constitute a gravel-sand body up to ~600 m wide and ~45 m thick. This body is deposited 232 within clay-dominated deposits. Clayey sediments overlying FA3 (Fig. 3) and recovered in 233 the uppermost borehole S164 (Fig. 2) were not described and interpreted in detail due to the 234 cursory description of Cerha (1987). 235 The sedimentary architecture of the Hrušky Formation can only be studied in FA3 236 based on two historical photographs of the Czujan’s sandpit profiles taken by R. Musil in 237 1964. The first profile shows co-sets of sand and gravelly sand with trough cross-stratification 238 (St, SGt; Fig. 4A) and co-sets of sand and gravelly sand with horizontal stratification (Sh, 239 SGh; Fig. 4A). Strongly eroded bases are recognized in the co-sets of facies (St, SGt) and the 240 individual sets (red and blue lines respectively; Fig. 4A). Some sections show the cross strata 241 onlap to the rising bases (white arrows; Fig. 4A). The individual sets (blue lines, Fig. 4A) and 242 the co-sets of facies Sh, SGh (green lines; Fig. 4A) have either subhorizontally or planar 243 bases. In contrast, the second profile (Fig. 4B) shows a sequence of two sedimentary units. 244 The lower unit is formed by co-set of planar cross-stratified sand (Sp facies), where individual 245 sets are separated by either planar or variously inclined surfaces (blue lines; Fig 4B). A large 246 body of clay (yellow arrow) is discerned in facies Sp (Fig 4B). This lenticular clay body was 247 originaly placed horizontally in the profile, but in fact it is partially deformed due to a partial 10 248 sliding of the sandpit wall (Fig 4B). In turn, the upper unit is formed by co-sets of sand and 249 gravel with low-angle cross stratification (co-sets of facies Sl, Gl; Fig 4B), where the bases of 250 the sets are faintly scoured (red lines; Fig 4B). 251 252 5.2 The faunal assemblage from Czujan’s sandpit 253 The following sections report on the fossil vertebrates from Czujan’s sandpit (see 254 updated faunal list with MNI in Table 1). 255 256 5.2.1 Reptiles 257 The herpetological assemblage only includes nine specimens of Testudines, of which 258 one belongs to a soft-shelled turtle and the remaining ones to a middle-sized testudinid. 259 PIUW-8X1939 (Fig. 5A) is a costal plate fragment, as shown by the presence of a rib on the 260 visceral part. Dorsally, the distinctive sculpturing consisting of small rounded pits, separated 261 from one another, allows us to refer the specimen to Trionychidae indet. (Marmi and Luján, 262 2012; Vitek and Joyce, 2015). Thenius (1951) reported from Czujan’s sandpit the presence of 263 Trionyx sp. However, given that the sculpturing pattern does not allow distinguishing between 264 the two genera of soft-shelled turtles recorded from the Miocene of Central Europe (i.e., 265 Rafetus and Trionyx; Georgalis and Joyce, 2017), identification at the genus rank is not 266 possible. 267 The remaining Testudines remains belong to a medium-sized testudinid, which was 268 reported as Testudo sp. by Thenius (1951). Four specimens have three-dimensionally 269 preserved carapace. The most complete shell (MZM Ot7877; Fig. 5B–C) is slightly elongated 270 and preserves the xiphiplastra (Fig. 5C), which indicates the lack of a hypo-xiphiplastral hinge 271 (Delfino et al., 2012). The peripherals 1–3 are not crossed by the pleuromarginal sulcus, 272 unlike in geoemydids (Luján et al., 2014). Peripherals 3–7 are completely involved in the 11 273 shell bridge and not elevated relative to overall shell height. Peripherals 8–10 are well 274 developed posteroventrally (Figs. 5B, 6G), but not curved medially as in Testudo canetoniana 275 (see Lapparent de Broin, 2000). The complete pygal shows that the supracaudal scute is not 276 divided by a sagittal groove. The anterior plastral lobe is trapezoidal and the anterior edge is 277 truncated (it does not protrude from the carapace contour). The neural 1 is subrectangular and 278 not constricted posteriorly. Based on these features, the Testudo specimens from Czujan’s 279 sandpit are referred to Testudo (Chersine) kalksburgensis, which is only known from its type 280 locality (Wien-Kalksburg, Vienna Basin), dated to the Middle Miocene (MN5–MN6; Toula, 281 1896; Bachmayer and Młynarski, 1981; Luján, 2015; Luján et al., 2016, in press). 282 283 5.2.2 Mammals 284 Proboscidean remains belong to three genera from different families — 285 Prodeinotherium (Deinotheriidae), Zygolophodon (Mammutidae), and Gomphotherium 286 (Gomphotheriidae) — of which only the last one was mentioned by Thenius (1951). 287 Deinotheres are only represented by a single M3 (MZM Ot7526; Fig. 5D), which displays a 288 bilophodont occlusal pattern with a distally tapering contour, a moderately developed distal 289 cingulum, and a distinct convolute (postmetaloph ornamentation) close to the distal cingulum 290 at about crown midline. The small dimensions of this tooth (62 cm in length and 59 cm in 291 width) overlap with Prodeinotherium (Ginsburg and Chevrier, 2001; Gasamans et al., in 292 press) and enable to rule out the assignment to a species of Deinotherium. Although the two 293 European species of Prodeinotherium cannot be distinguished based on M3 morphology 294 (Ginsburg and Chevrier, 2001; Gasamans et al., in press), Prodeinotherium cuvieri is only 295 recorded from the early Miocene (Ginsburg and Chevrier, 2001; Böhme et al., 2012; Pickford 296 and Purabrishami, 2013; Gasamans et al., in press) so that a tentative assignment to 297 Prodeinotherium cf. bavaricum is warranted. 12 298 In turn, mammutid remains are the most abundant, being represented by dental as well 299 as cranial and postcranial remains. This is very exceptional for Mammutidae, which is much 300 less common than Deinotheriidae and Gomphotheriidae in the European Miocene (see 301 Göhlich, 2010). We concur with previous authors (Holec, 1985; Seitl, 1985) that 302 Zygolophodon turicensis is the only mammutid species recorded at Czujan’s sandpit, as 303 supported by the morphological features displayed by the M3 (MZM Ot7519; Fig. 5E), 304 namely mesiodistally compressed lophs with well defined zygodont crests, long interlophs, 305 well-developed cingula, a disctinctly lower and narrower fourth loph, and a rhombic wear 306 pattern in the pretritte (Tobien, 1975, 1996; Tassy, 1977). Finally, only an M2 (MZM Ot7524; 307 Fig. 5F) and an upper tusk fragment (MZM Ot7505; Fig. 5G) testify to the presence of a 308 gomphothere. The M2 crown is trilophodont and displays a typical bunodont pattern (Tobien, 309 1973), with relatively thick enamel, an asymmetrical trefoil wear pattern in the pretritte, a 310 reduced lingual cingulum, massive cusps, and narrow interlophs that are blocked by enlarged 311 central conules. These characters allow us to discount the assignment of the material to the 312 early diverging species of the Gomphotherium (annectens) species group, and further 313 distinguish the Czujan’s M2 from the subtapiroid dental pattern of the more derived species 314 Gomphotherium subtapiroideum and Gomphotherium steinheimense (see Gohlich, 2010; 315 Tassy, 2014). The tusk fragment has a pyriform cross section and is twisted, with a 316 helicoidally oriented enamel band. The latter morphology is characteristic of G. angustidens, 317 whereas the upper tusks of G. subtapiroideum, G. steinheimense, and Archaeobelodon filholi 318 lack torsion (Göhlich, 1998, 2010; Tassy, 2014). We therefore concur with Seitl (1985) that 319 the Czujan’s gomphothere is attributable to G. angustidens. 320 Three perissodactyl families are represented at Czujan’s sandpit: Chalicotheriidae, 321 Rhinocerotidae, and Equidae. Thenius (1951) already reported the presence of 322 Chalicotherium grande (currently in Anisodon; Anquetin et al., 2007), but the original 13 323 material has not been found. The currently available most complete material is represented by 324 a fourth metacarpal (MZM Ot7766; Fig. 5H). It displays a roundish distal facet and a deep 325 lateral ligament fossa, and the shaft is lateromedially compressed but widens distally. The 326 dorsal side of the shaft is straight, whereas the palmar side is concave. The concave palmar 327 side and the presence of a single facet for metacarpal III in the Czujan specimen distinguish it 328 from the morphology displayed by the schizotheriines, which display a more massive and 329 dorsoventrally compressed metacarpal IV with two separate facets for the metacarpal III 330 (Zapfe, 1979). MZM Ot7766 fits well in size and shape with the remains of Anisodon grande 331 from Devínská Nová Ves - Zapfe´s fissure (Zapfe, 1979) and Sansan (Guérin, 2012), whereas 332 Chalicotherium goldfussi shows slightly larger dimensions (Guérin, 2012). 333 Two rhinocerotid genera are recognized at Czujan’s sandpit on the basis of different 334 limb bone proportions: Hoploaceratherium and Brachypotherium. The right complete tibia 335 MZM Ot7749 (Fig. 5I) displays more elongated proportions than in the teleoceratin 336 Brachypotherium (Fig. 5J) and compared with other long-legged rhinocerotid genera, it more 337 closely matches the morphology of Hoploaceratherium (Heissig, 2009, 2012). The shaft is 338 slenderer than in Aceratherium and the distal epiphysis is broader than in Lartetotherium 339 (Hünermann, 1989; Heissig, 2012). The caudal plane below the condyles is medially bounded 340 by a sharp ridge, and the frontal groove of tuberositas tibiae is narrow and centrally located. 341 These characters are typical for Aceratheriini and do not occur in Lartetotherium (Heissig, 342 2009, 2012). 343 In turn, the genus Brachypotherium is represented by dentognathic (an upper incisor 344 and a maxillary fragment) and postcranial (tibia) remains. The slightly worn I1 (MZM 345 Ot7700; Fig 5K) displays a short blunt root and differs from those of Aceratherium and 346 Lartetotherium, which are somewhat smaller and possess a pointed root (Heissig, 2012). In 347 turn, an assignment to Hoploaceratherium can be excluded due to the lack of upper incisors in 14 348 the latter genus (Heissig, 2012). The M2–M3 from the maxillary fragment MZM Ot7752 (Fig. 349 5L) display a brachyodont morphology, with a strong lingual cingulum surrounding the 350 protocone and extending to the base of the hypocone; the M2 has a weak buccal cingulum and 351 a well-developed crochet. These molars differ from those of Aceratheriini and Rhinocerotini 352 by their greater dimensions and stronger cingulum (Heissig, 2012). The complete left tibia 353 MZM Ot7752 (Fig. 5J) is relatively shorter and more robust than in the “long-legged” 354 Aceratheriini and Rhinocerotini (Cerdeño, 1993; Heissig, 2012), supporting an assignment to 355 Brachypotherium. The dimensions of M3 and tibia correspond well to those of 356 Brachypotherium brachypus from Malartic (MN7+8, France; Cerdeño, 1993), in agreement 357 with the previous report by Thenius (1951) of B. cf. brachypus from Czujan’s sandpit based 358 on unknown material. 359 As for the equids, only an anchitheriine maxillary fragment with fragmentary M1 and 360 very worn M2 is available (MZM Ot7765; Fig. 5M). The molars are brachyodont and broader 361 than long; the metaloph and protoloph hooked in distal direction, and only a vestigial lingual 362 cingulum is present. Three anchitheriine species have been recorded from Central Europe: 363 Anchitherium aurelianense, Anchitherium hippoides, and Anchitherium steinheimense 364 (Abusch-Seiwert, 1983; Rotgers et al., 2011). However, we refer MZM Ot7765 to 365 Anchitherium sp. because the advanced degree of wear makes it impossible to ascertain 366 occlusal details. 367 Artiodactyls are represented at Czujan’s sandpit by five taxa from four different 368 families: Bovidae, Palaeomerycidae, Cervidae and Suidae. The bovid material includes a 369 single frontoparietal with horn cores (MZM Ot7786; Fig. 5N–O). The horn cores are long, 370 straight, and moderately divergent, and display a mediolateral uniform compression and weak 371 longitudinal grooves from base to apex; the pedicels are tall (Fig. 5N–O). MZM Ot7786 372 represents the most complete material and fits well in size and shape with the material from 15 373 Czujan’s sandpit and Klein-hadersdorf used by Thenius (1951) to erect “Gazella” stehlini and 374 subsequently assigned to ?Tethytragus stehlini by Azanza and Morales (1994; see also Bibi 375 and Güleç, 2008). We note that MZM Ot7786 is more complete than the holotype (only a 376 gypsum cast was inspected in PIUW) of Te. stehlini, which is represented by a single left horn 377 core fragment. 378 A palaeomerycid mandibular fragment with right m2–m3 is preserved (MZM Ot7773; 379 Fig. 5P). The lower molars display a rugose enamel surface (particularly on the lingual side) 380 and a brachyselenodont morphology with strong deep valleys and pointed conids (Fig. 5P). 381 This combination of features supports the attribution to Palaeomerycidae indet., but does not 382 enable a more refined assignment to a genus due to the absence of more complete cranial 383 material. 384 Three cervid antler fragments with consistent morphology are preserved. MZM Ot7771 385 (Fig. 5Q) is the most complete cranial appendage, where only the anterior branch and 386 proximal part of the pedicle are broken off. The antler is straight, laterally compressed and 387 bifurcated, without burr, but with extended, strongly sculptered base with longitudinal 388 grooves and ridges. The studied antlers clearly differ from coronate antlers of Lagomeryx and 389 Paradicrocerus (previously Stehlinoceros, see Böhme et al., 2012). MZM Ot7771 is 390 dichotomous and more closely resembles the antlers of Heteroprox, being distinguished from 391 other European genera by the absence of a burr (e.g., Heckeberg, 2017; Rössner et al., 2021). 392 The general morphology and dimensions of MZM Ot7771 only fit well with those of 393 Heteroprox, which is known by three species in Europe. The appendage allows an assignment 394 to Heteroprox larteti by the larger dimensions and stronger sculpture than those of H. eggeri 395 (see Rössner, 2010). Its smaller dimensions, more ellipsoid cross section of the pedicle and 396 short anterior prong, excludes an assignment to Heteroprox moralesi (see Rössner, 2010). 397 Suids are represented by two mandibular fragments that, based on their similar 16 398 preservation, might belong to the same individual (Fig. 5R): a right fragment with m1–m2 399 (RMM P11a) and a left one with a partial m3 preserving the distal crown portion (RMM 400 P11b). The moderately worn m1 (16.5 x 10.4 mm) and the less worn and larger m2 (19.0 x 401 12.7 mm) display a similar morphology, with thick enamel, a similarly narrow subrectangular 402 occlusal outline, conspicuous furrows between the main cusps and the central and distal 403 pillars, a well developed mesiobuccal cingulum, and a distinct hypoectoconulid. The distal m3 404 fragment displays a relatively short talonid that is only slightly tilted buccally, with the 405 hypoconulid surrounded by cingular cusplets. This morphology resembles the tetraconodont 406 genera Conohyus and Retroporcus, which have been the subject of taxonomic revision during 407 the last decade (Pickford and Laurent, 2014; Pickford, 2016) and are still a matter of debate 408 (van der Made, 2020). Pickford and Laurent (2014) designated a lectotype for Conohyus 409 simorrensis and provided an emended diagnosis of the species, distinguishing the genus 410 Conohyus from the new genus Retroporcus (see also Pickford, 2016), which includes material 411 previously assigned to C. simorrensis by various authors (e.g., van der Made, 1989; van der 412 Made and Salesa, 2004). Pickford and Laurent (2014) and Pickford (2016) interpreted 413 Retroporcus matritensis as a senior synonym of C. simorrensis goeriachensis, whereas van 414 der Made (2020) questioned the lectotype designation by Pickford and Laurent (2014) and 415 considered R. matritensis a junior synonym of C. simorrensis. The suid mandibular fragments 416 from Czujan’s sandpit would be referrable to C. simorrensis sensu van der Made (2020), but 417 until Pickford and Laurent’s (2014) lectotype designation is proven wrong, we prefer to 418 follow their concept of C. simorrensis and distinguish R. matritensis as a distinct species. 419 Although both species largely overlap in the size of their teeth, the proportions of the m1 and 420 m2 from Czujan’s sandpit do not fit well with those of C. simorrensis sensu Pickford and 421 Laurent (2014), being relatively narrower and most closely resembling those of R. matritensis 422 (Pickford, 2013, 2016). The slightly tilted m3 talonid also more closely resembles the latter 17 423 species, since C. simorrensis is characterized by a less symmetrical talonid (Pickford and 424 Laurent, 2014; Pickford, 2016). Although additional material (particularly the premolar 425 series) would be required to confirm our identification, based on the aforementioned features 426 we tentatively assign the material to cf. R. matritensis. Similarly to C. simorrensis, R. 427 matritensis is known from European localities ranging from MN5 to MN9 (Pickford and 428 Laurent, 2014; Pickford, 2016). 429 The carnivoran remains from Czujan’s sandpit consist only of two partial lower canines 430 (ÚGV Pal297, Fig. 5S; MZM Ot7814) and a calcaneum (MZM Ot7815; Fig. 5T). The more 431 complete canine (>8.2 cm in height) preserves the entire root and most of the crown (except 432 for the apex; ÚGV Pal297; Fig. 5S). The base of the crown is labiolingually compressed and 433 preserves a distinct wide lingual wear facet to accommodate the I3 (Fig. 5S). In mesiodistal 434 view, ÚGV Pal297 is slightly sigmoidal, which is typical for several groups of carnivorans. 435 According to Viranta (1996), two valid amphicyonid genera are known from the middle 436 Miocene of Europe: Megamphicyon giganteus (MN3–MN5), until recently included in 437 Amphicyon (Siliceo et al., 2020); and Amphicyon major (MN4–MN9). The canine root of 438 ÚGV Pal297 is less robust than in Megamphicyon giganteus (see Kuss, 1965). Based on size 439 and root proportions, both canines are thus referred to Amphicyon (Ginsburg, 1961; Ginsburg 440 and Antunes, 1968). The calcaneum MZM Ot7815 (Fig. 5S) is slightly abraded, especially the 441 tuber calcanei, but it preserves enough diagnostic features to further support the assignment to 442 Amphicyon. It is relatively short (8 cm long) and displays a robust tuber calcanei. The ectal 443 facet is convex, craniocaudally elongated, and completely separated from the sustentacular 444 facet. The latter is rather flat, semicircular in outline, and medially protruding. The referral to 445 Amphicyon is based on size as well as the robustness of the tuber calcanei and the shape of the 446 ectal facet (Argot, 2010). Given the restricted available material, we only tentatively assign 447 the material to Am. cf. major. 18 448 449 5.3 Taphonomy 450 Despite the lack of field documentation, many fragmentary remains could be joined 451 into complete bones during material preparation. This fact, and the presence of fresh fractures 452 on the bone fragments, indicate that they were well-preserved and generally complete in the 453 field, but that they were not adequately collected due to the “rescue” character of the 454 excavations. Bones and teeth without any sign of transportation include 91% of the studied 455 material, and 31% of the studied material represents complete skeletal elements from various 456 parts of skeletons: teeth, skull fragments, mandibles, vertebrae, ribs, scapula, humerus, ulna, 457 radius, and autopodial bones. The preservation of the vertebrate remains from Czujan’s 458 sandpit may be summarized as follows: 1) there is a high number of complete bones and teeth, 459 being the former more frequent than the latter; 2) teeth are preserved both as dentognathic 460 fragments (Fig. 5G, L, M, P, R) and isolated tooth crowns with roots (e.g., Fig. 5D, E, F, K, 461 S); 3) we identified up to 29 cases of antimere bones (e.g., right and left humerus: Fig. 6A–B) 462 and teeth; 4) in 10 cases, different bones appear to belong to the same individual and have 463 been preserved in articulation or close spatial association, based on anatomical congruence, 464 preservation state and/or information from labels (Fig. 6C–E), while in four cases unfused 465 epiphyses matched with the corresponding diaphysis (Fig. 6F); 5) as pointed out by Jüttner 466 (1938), there were well-preserved proboscidean skulls in Czujan’s sand pit (either not 467 recovered or subsequently disintegrated); 6) four of the seven testudinid specimens preserved 468 well their 3D morphology and growth scute lines are discernible in some carapace plates 469 (Figs. 5B, 6G). The presence of antimeres and articulated specimens among the proboscidean 470 sample, coupled with similar preservation, indicates that most of the bones come from a 471 single individual—as further supported by information from the museum label of the 472 proboscidean atlas RMM P34, according to which it was articulated with the skull in situ. 19 473 Generally, all bones and teeth are strongly mineralized in a similar way (Březina, 474 2019), and the mineralization of the more rounded and abraded bones is comparable with 475 most of the better preserved ones, suggesting the lack of reworking and resedimentation of 476 fossils from older deposits. Bone cavities are filled by quartzose sands, gravels, and clays. 477 Finally, some specimens (e.g., MZM Ot7497 and MZM Ot7694) may contain remnants of soil 478 or plant roots. Most of the fossil bones are fresh and angular (Figs. 5B–C, G, H–J, N–O; 6A– 479 D, G-H) and only rarely are deformed (Fig. 6I). Prediagenetic traces of bone corrosion are 480 infrequent. Only weathering stage 1 (sensu Behrensmeyer, 1978) modifications were 481 observed, although longitudinal cracks in diaphyses and articular facets are often present (Fig. 482 6I–J). Two gnawing traces, probably produced by some large carnivores (Fig. 6K–L) and a 483 rodent, were identified (Fig. 6M). However, we cannot exclude that some gnawing traces 484 were alternatively produced by a ruminant, as they show a similar pattern to those produced 485 by rodents (Hutson et al., 2013). 486 Most of the studied material belongs to terrestrial vertebrates and shows no signs of 487 redeposition. In contrast, rare marine fossils indicate redeposition (Seitl 1985). In particular, 488 the analyzed pelitic matrix taken from the fossil bones only sporadically contains microfaunal 489 marine remains, including: a single placoid shark scale; and the following foraminifera 490 (Arenobulimina sp., Amonodiscus cf. glabratus, Paragloborotalia acrostoma, Heteroleppa 491 dutemplei, and Globigerinida indet.), which must have been redeposited from the lower 492 Badenian Ždánice Unit (R. Brzobohatý and M. Bubík, pers. comm. to J.B.) as well as a single 493 isolated shark tooth crown (Fig. 6N) referred to Isurus sp. The macroflora is represented by 494 non-redeposited scarce wood remains (Fig. 6O), whereas palynomorphs were not present in 495 the analyzed sediment samples. 496 All recovered testudinids from Czujan’s sandpit are adult individuals, not only based 497 on size, but further because the sutures between plates are completely fused or poorly visible 20 498 in some cases. Based on the third molar eruption and/or epiphyseal fusion, mammals from 499 Czujan’s sandpit are adult individuals, with the only exception of proboscideans. Ontogenetic 500 analysis, based on dental wear stages in the proboscidean samples (see Tassy, 1996, 2013), 501 revealed different ontogenetic stages, including three juveniles, two mature subadults and 502 fifteen midlife adults (Table. 2). 503 504 6. Discussion 505 6.1 Sedimentological interpretation 506 From a sedimentological viewpoint (Figs. 3–4), the original outcrops of Czujan’s 507 sandpit were formed by tabular bars (Sh, SGh) and channel infills (St, SGt, Sp), which fill the 508 distributary channels of both deltaic plain and deltaic front of a shallow water fluvial 509 dominated delta (Postma, 1990; Chen et al., 2015; Zhang et al., 2017). The onlaps of cross 510 strata arise during migration of climbing dunes in mouth bars of a delta front (Dasgupta et al., 511 2016), whereas the superposition of co-set facies Sl, Gl over co-set facies Sp can be 512 interpreted as delta plain deposits overlapping mouth bar sediments (Francírek, 2018). The 513 clay body (Fig. 4B) is most probably interdistributary bay infill (Zhu et al., 2017). As is 514 typical for braid deltas, the delta plain is composed only by distributary channel infills 515 (McPherson et al., 1988). Both the coarsening-upward sequence of facies association FA1– 516 FA3 and the lens shape of megascale body FA2+FA3 reflect a delta progradation into a 517 marine basin. According to this interpretation, three facies associations are recognized: FA1, 518 composed by pelagic sediments; FA2, prodelta sediments (suspension clays plus occasional 519 sand turbidites) up to the delta slope sediments (sand turbidites plus suspension sediments of 520 quiet phase on delta slope); FA3, sediments of delta front (mounth bars and interdistributary 521 bays) and delta plain (subaerial distributary channels) infills. The latter interpretation is 522 supported by the occurrence of wood fragments (Fig. 6O), which are common in distributory 21 523 channels or their mouth along the delta front (Olariu and Bhattacharya, 2006), as well as by 524 the presence of terrestrial fauna. Based on the sedimentary architecture, Czujan’s sandpit can 525 be interpreted as a shallow-water river delta with braid delta plain and with mouth bars in 526 distributory outlets and interdistributary bays on delta front (Nemec, 1990; Postma, 1990; Zhu 527 et al., 2017). 528 529 6.2 Taphonomical interpretation 530 The bone concentration from Czujan’s sandpit is characterized by the relatively diverse 531 assemblage of terrestrial vertebrates, the presence of complete and associated bones (skull 532 bones, mandibles, vertebrae, ribs, limb bones), a low degree of weathering, and a 533 predominance of fresh and angular specimens. These aspects, and the absence of polished or 534 otherwise discrepant bone of terrestrial vertebrates characteristic for exhumation or 535 redepositions (Rogers and Kidwell, 2007), unequivocally indicates that the terrestrial 536 vertebrate assemblage is not mixed with reworked older terrestrial fossils. In contrast, the rare 537 marine fossils represent redepositions from the Ždánice Unit, Lanžhot Formation and lower 538 parts of Hrušky Formation (Seitl 1985). Reworked, allochthonous marine fossils have been 539 already documented by Kuklová (1970, 1978) from the Wielician-Kosovian clays in 540 boreholes (Figs. 2, 3). 541 The presence of bones and teeth from more than a single individual and with multiple 542 species represented allows us to classify at least the sands and gravels of FA3 (vertebrate 543 remains are not documented from sands and gravels of FA1–FA2) as a macrofossil multitaxic 544 bonebed (Behrensmeyer, 2007; Eberth et al., 2007; Rogers and Kidwell, 2007). Because 545 mining extraction in Czujan’s sandpit from the 1930s to the 1980s progressed from south 546 towards north and east, and abandoned parts of the sandpit were gradually filled by waste as 547 extraction progressed, new and relatively small outcrops (in comparition with the whole 22 548 mining area) were accessible in each mining phase (Březina, 2019). This implies that more 549 than a one bone bed was present in the Czujan’s sandpit profiles. The different types of rocks 550 represented by the matrix adhered to the fossils as well as the presence of recent soil remnants 551 on some specimens suggest that the fossil remains came from different stratigraphic positions. 552 This is further supported by the old label of PIUW 4501–4503, which clearly describes the 553 find at 6.5 m depth, and corresponds well with Seitl’s (1985) observation that bones and teeth 554 were vertically scattered along the whole profile. Therefore, from a finer scale perspective, we 555 can conclude that the bone accumulations of FA3 were located in more than one chanel infill, 556 probably as a result of gradual or periodical fluvial sedimentation. In the absence of field 557 documentation, it is impossible to confidently establish whether the fossil bones originated 558 from a time-averaged accumulation of carcasses randomly transported from the river basin 559 and deposited due to the loss of water energy along the mouth of distributary channels into the 560 sea. Alternatively, the assemblage might have originated by the progressive accumulation 561 from a preexisting source of mass death accumulation. However, the high concentration of 562 terrestrial taxa is outstanding in the context of surrounding marine deposits, and the low 563 number of carnivorans and the presence of a single specimen of a freshwater taxon 564 (trionychid) rule out either a flooding event or a miring mortality (Rogers and Kidwell, 2007). 565 Moreover, multitaxic fossil assemblages, formed by immature individuals as well as gnawing 566 traces, do not occur in mass drowning (Rogers and Kidwell, 2007; Backwell et al., 2018). 567 Both the presence of cracks and the gnawing traces suggest that some carcasses were laying 568 for some time on the surface before sediment deposition. 569 The main taphonomic features of the Czujan’s sandpit assemblage (i.e., wearing stage 570 1 sensu Behrensmeyer, 1978, a minimal number of gnawed bones, and evidence for 571 articulated skeletons) are often reported from drought mass death sites (Haynes, 1985, 1988, 572 1991, 2018, Backwell et al., 2018). The structure and MNI of taxa such as proboscideans, 23 573 bovids or testudines from Czujan’s sandpit (Table 1) further resemble those from extant sites 574 from Africa, where drought mass death implies a higher mortality of proboscideans and 575 bovids (Haynes, 1988). Predominance of large herbivores such as proboscideans (Table 1), 576 which are characterized by high mortality during dry periods, might support this idea 577 (Haynes, 1985, 1991, 2017; Haynes and Klimowicz, 2015). Variously old proboscidean 578 individuals have been recovered from Czujan’s sandpit fossil record, including juveniles 579 (Table 2), which are characteristic in recent drought mass death sites (Haynes, 1985, 1991, 580 2017; Haynes and Klimowicz, 2015). Based on our findings, a progressive secondary 581 accumulation from preexisting drought mass death accumulations into FA3 seems most 582 likely. This interpretation would be in agreement with the strong decrease of humidity in 583 Central Europe during the late Badenian, as well as the presence of seasonality in the Vienna 584 Basin accompanied by periodic drought conditions (e.g., Böhme et al., 2011; Harzhauser et 585 al., 2011), supported by documented braid delta. This delta type occurs in areas with arid 586 climate and limited vegetation, fed by flows with unbalanced flow rate (McPherson et al., 587 1988; Miall, 1996). However, poor field data do not allow to clearly determine if the fossil 588 assemblage is a time-averaged abiotic accumulation or whether it was deposited following 589 drought mass death accumulation. Taking in account the dispersal capabilities of large 590 mammals, the studied taphocoenosis probably mixes taxa from various biotopes concentrated 591 in one place of the river basin for both possible intepretations. In the analogy with recent 592 elephant populations whose home ranges reach up to ~3000 km2 (Haynes, 1991), we assume 593 that animals might have inhabited area from the Vienna Basin coastline, including delta 594 platform to the eastern slopes of Bohemian Masiff, which is largely formed by the 595 soutwestern part of the Carpathian Foredeep Basin (see Fig. 1). 596 597 6.3 Biostratigraphy 24 598 All mammal taxa identified in the Czujan’s sandpit have been reported from the 599 European early Astaracian (MN6) localities. Although the mammal community as a whole 600 displays a wide biostratigraphic range from MN3 up to MN10, the first occurrence of R. 601 matritensis and Hoploaceratherium in Göriach, Austria (late MN5, ⁓14.5 Ma; Aiglstorfer et 602 al., 2014), and Te. stehlini in Klein-hadersdorf, Austria (MN5-MN6, ⁓14.2 Ma; sensu Böhme 603 et al., 2012) together with the last known occurrence of P. bavaricum from Devínska Nová 604 Ves-Zapfe’s fissures, Slovakia (early MN6; Fejfar 1990) restrict the biostratigraphic age of 605 the Czujan’s sandpit assemblage within the late MN5 to late MN6 Zones (Fig. 7). Given by 606 the fossil record of mammals from well-dated Central European localities, the above 607 mentioned biostratigraphic age is supported by the coeval first appearance of R. matritensis 608 and Te. stehlini and the last appearance of P. bavaricum, Retroporcus matritensis, H. larteti 609 and Am. major (Fig. 7). The Badenian age of the vertebrate assemblage is also partially 610 supported by the presence of a testudinid T. kalksburgensis reported from Wien-Kalksburg, 611 Austria (MN5–MN6; Bachmayer and Młynarski, 1981). 612 The biostratigraphical age (late MN5, to late MN6; late early to late Badenian; Fig. 7), 613 based exclusively on the vertebrate assemblage, is generally in agreement with the late 614 Badenian age of the Czujan’s sandpit, as previously suggested based on marine and brackish 615 microfauna (Kosovian: Buday et al., 1964; Čtyroký, 1989; Čtyroký et al., 1990; Fejfar, 1990 616 and Wielician: Kuklová, 1978; Stráník et al., 1999; Jiříček, 2002). These authors did not 617 verify the age on the basis of vertebrate fauna, such as Thenius (1951) and Musil (1956) and 618 Seitl (1985), who adopted Jüttner's (1938, 1940) exclusively petrographic correlation with the 619 Sarmatian sands. However, Sarmatian age is highly improbable, because Sarmatian sediments 620 have not been documented from the Mikulov area (Fig. 1B). The presence of the Ammonia 621 beccarii ecozone, laterally replaced by the Bulimina-Bolivina Zone in the Nové Mlýny-2 622 borehole (Bimka, 1983), proves that both FA3 and FA2, together with the uppermost section 25 623 of FA1, correspond to the Kosovian substage (Figs. 2–3). The lower limit of the Kosovian 624 substage varies within the range of 13.6–13.1 Ma (Hohenegger et al., 2014) and therefore, 625 Czujan’s sandpit assemblage cannot be older than ~13.6 Ma. This corresponds to the upper 626 MN6 boundary in Central Europe (base of C5ABn sensu Steininger, 1999, but see different 627 MN Zonation in Western Europe, Fig. 7). 628 According to the Nové Mlýny 2 borehole (Bimka, 1983), the section of Kosovian 629 continues minimally for at least ~40 m above the sandpit, however, the original thickness of 630 Kosovian section was most probably greater but subsequently reduced by Sarmatian and 631 Quarternary erosions. Therefore, both the close proximity to the Wielician/Kosovian 632 boundary (Fig. 2, 3) and mammal biochronology (Fig. 7) support latest MN6 (~13.6 Ma) age 633 of the Czujan’s sandpit. Prodeinotherium and possibly also Am. major from Czujan’s sandpit 634 are among the youngest records of these taxa from Central Europe (late MN6, Kosovian 635 substage; Fig. 7). 636 637 6.4 Paleoecological implications 638 The vertebrate assemblage from Czujan’s sandpit comes from various paleohabitats. A 639 closed woodland habitat is indicated by proboscideans (Prodeinotherium cf. bavaricum and 640 Zygolophodon turicensis), rhinos (Brachypotherium brachypus and Holoaceratherium sp.), 641 chalicotheres (Anisodon grande), palaeomerycids, suids (Retroporcus matritensis) and 642 carnivorans (Amphicyon cf. major; Fig. 8). The dental morphology and the still rather limited 643 number of δ13C values from tooth enamel of P. bavaricum (Seegraben; Aiglstorfer et al., 644 2014) and B. brachypus (Steinheim; Tütken et al., 2006) indicate that both species might have 645 inhabited a closed woodland environment. Mammutids of the genus Zygolophodon browsed 646 in forest habitats as well (e.g., Lambert and Shoshani, 1998) but based on zygolophodont 647 deintition their dietary spectrum was most probably different from that of bunodont 26 648 gomphotheres and lophodont deinotheres. This assumption is indirectly supported by feeding 649 preferences of conifers (Taxodium) in closely related American mastodons (Mammut 650 americanum) from the Pleistocene of Florida (Green et al., 2005), where Taxodium swamps 651 are comparable to those documented from the Miocene of Central Europe (e.g., Kvaček et al., 652 2004; Kázmér, 2007). The brachyodont teeth of palaeomerycids, as well as their isotopic data 653 from middle Miocene localities of Germany and Spain, support a folivorous diet in woodlands 654 (Tütken et al., 2006; Tütken and Vennemann, 2009; Domingo et al., 2012). A woodland 655 habitat is also favored by the chalicotheriid A. grande, which is considered a specialized 656 browser (Schulz et al., 2007), as well as the amphicyonid Am. major, which was an active 657 omnivorous predator (Argot, 2010) anatomically well-adapted for climbing trees and chasing 658 preys (Argot, 2010). 659 The presence of mixed-feeders such as Gomphotherium, Anchitherium, Heteroprox 660 and Tethytragus suggest, in addition, the existence of more open environments. The genus 661 Gomphotherium has been considered an inhabitant of open environments similar to recent 662 African savannas (Tassy, 1977; Lambert, 1996; Lambert and Shoshani, 1998). The isotopic 663 data for subtapiroid species of Gomphotherium from the latest early and early middle 664 Miocene of Germany suggests the consumption of C3 vegetation in woodland environments, 665 as documented by δ13C values in G. steinheimense (‒10.7‰ up to ‒10.1‰; Tütken et al., 666 2006) and G. subtapiroideum (‒11.9‰ to ‒10.2‰; Tütken and Vennemanm, 2009). However, 667 dental microwear studies of both species indicate a mixed-feeding, with G. subtapiroideum 668 likely favoring more open woodlands (Calandra et al., 2008, 2010). In contrast, the rather high 669 δ13C values for G. angustidens from the middle Miocene localities of the Madrid Basin, Spain 670 (‒10.65‰ to ‒6.21‰; Domingo et al., 2009, 2012) indicate that Gomphotherium might have 671 intruded into open and grassland environments. Although data from the Central European 672 populations of G. angustidens are missing, this species most probably favored more open 27 673 environments than the other proboscidean species recorded from Czujan’s sandpit (i.e., P. 674 bavaricum and Z. turicensis), which most probably occupied a different ecological niche in 675 more closed woodland (Fig. 8). Anchitherium might also have inhabited open woodland 676 environments, as suggested by results from dental mesowear (Kaiser, 2009) and isotopic 677 studies (Tütken et al., 2006; Tütken and Vennemann, 2009) of A. aurelianense populations 678 from Sandelzhausen and Steinheim (Germany), as well as more open environments as 679 suggested by δ13C values (‒12.7‰ to ‒6.49‰; Domingo et al., 2012) from populations of 680 Spanish MN5‒MN6 localities. 681 Mixed-feeding Heteroprox species usually inhabited forests or woodlands in Central 682 Europe (specifically Seegraben, Austria and Sandelzhausen, Germany) during the Langhian, 683 as documented by δ13C values (‒12.0‰ to 10.4‰; Tütken and Vennemann, 2009; Aiglstorfer 684 et al., 2014). However, the dental microwear study of H. larteti populations from central 685 Spain (DeMiguel et al., 2011), contrary to isotopic studies of Domingo et al. (2012), revealed 686 a high degree of grazing interpreted as an adaptation to seasonal arid periods during the 687 Serravallian. A similar variation in paleohabitat is recorded in Tethytragus. Isotope data of 688 Tethytragus sp. from Gratkorn, Austria (MN7+8; Aiglstorfer et al., 2014) indicate a woodland 689 environment, whereas isotope data of Tethytragus langai from Paracuellos 3, Spain (MN6; 690 Domingo et al., 2012) show a mixed-feeding diet related to more open conditions. Augustí 691 and Antón (2002) pointed out that Tethytragus was capable of inhabiting different habitats 692 with a diet comprising a variety of vegetation. This was confirmed by micro- and mesowear 693 analysis of Tethytragus langai from the Serravallian of Spain (DeMiguel et al., 2011). Hence, 694 seasonal variations of dietary composition in Central European populations (e.g., those from 695 Gratkorn) could have occurred as well (Aiglstorfer et al., 2014). The omnivorous Retroporcus 696 matritensis also preferred a more open woodland environment based on isotopic studies 697 (Tütken et al., 2006; Domingo et al., 2009). 28 698 Among rhinocerotids, Brachypotherium brachypus has been traditionally considered a 699 semiaquatic species of open environments (Heissig, 1999; Augustí and Antón, 2002; Costeur 700 et al., 2012). However, its brachyodont dentition, and the isotopical data of Brachypotherium 701 from Steinheim (MN7, Germany) and Eichkogel (MN11, Austria), suggest browsing in closed 702 woodland environment (Tütken et al., 2006; Aiglstorfer et al., 2014) unlike 703 Hoploaceratherium, which most likely favored more open environments based on δ13C values 704 from the middle Miocene (MN5–MN6) localities from the Madrid Basin, Spain (‒11.15‰ to 705 ‒9.02‰; Domingo et al., 2012). The different paleoecological preferences of these two 706 rhinocerotids are as well manifested with their anatomical differences, as Hoploaceratherium 707 has long legs and is better adaptated for a cursorial locomotion in an open landscape, whereas 708 B. brachypus is a short-limbed and hippo-like species (Cerdeño, 1993; Heissig, 2012). 709 The indeterminate trionychid turtle is indicative of the presence of permanent 710 freshwater reservoirs, whereas the testudinid Testudo kalksburgensis inhabited open 711 woodland, or even more open environments. Medium-sized testudinids are well adapted to 712 dry conditons, and can inhabit a wide variety of open environments ranging from clearings 713 with low vegetation to semiarid habitats (Miklas-Tempfer, 2003, 2005; Čerňanský et al., 714 2012; Luján et al., in press). The aquatic trionychid turtle, semiaquatic Brachypotherium 715 together with the rich fossil record of proboscideans at Czujan's sandpit, (whose recent 716 representatives are well known for water seeking; e.g., Haynes, 1991), further indicates the 717 presence of permanent freshwater bodies occurring in the northwest area of the Vienna Basin 718 coastline to the adjacent Carpathian Foredeep Basin. Middle Miocene changes in the Central 719 Paratethyan terrestrial environments were conditioned by increased tectonic activity which 720 resulted in uplifting mountains in the east and two marine transgressions, one during the early 721 Badenian and second one in the late Badenian (Kováč et al., 2007). The extensive areas 722 around the Vienna Basin were divided by the uplift of its eastern part into western lowlands 29 723 and eastern steep highlands in the late Badenian (e.g., Kvaček et al., 2006; Kováč et al., 724 2007). The late Badenian paleobotanical record from eastern Central Europe indicate that the 725 highland situated southeast of the Vienna Basin was covered by a mixed mesophytic forest 726 with extrazonal (mountain) vegetation (i.e., Devínska Nová Ves, Slovakia; Kvaček et al., 727 2006; Kováčová et al., 2011; Kovar-Eder and Teodoridis, 2018). The dominance of forest 728 environments is also documented in this area by the late Badenian mammal assemblages from 729 Devínska Nová Ves locality - Zapfe’s Fissures, “Bonanza” and Sandberg (Sabol and Kováč, 730 2006). Although paleobotanical record from lowlands situated in the Carpathian Foredeep 731 Basin, (i.e., northwest of the Vienna Basin) is rather incomplete, recent analyses from the 732 early late Badenian to early Sarmatian low altitude (0-300 m a.s.l.) localities of the Carpathian 733 Foredeep Basin, together with the data obtained both south and west of the Vienna Basin, 734 point to the presence of subhumid sclerophyllous vegetation (Kovar-Eder and Theodoridis, 735 2018). The paleoecological analysis of the late Badenian Czujan’s sandpit revealed a mosaic 736 of habitats represented by forests, close to open woodlands, more open environments, and 737 stagnant or slowly flowing freshwater reservoirs in the continental northwestern area of the 738 Vienna Basin to the adjacent Carpathian Foredeep Basin (Fig. 8). This mosaic of local 739 habitats could be understood as restricted riparian and floodplain forests surrounded by 740 widespread woodland which opened inland, locally passed into the more open environments 741 (grass dominated). 742 Subhumid sclerophyllous forests documented from the Central Paratethys suggest a 743 more marked precipitation seasonality for the late Langhian to early Serravalian period 744 (Kovar-Eder and Theodoridis, 2018). The predominance of open habitats, both west of the 745 Vienna Basin and in the southwestern part of the Carpatian Foredeep Basin (see also section 746 5.3.; table 1 and 2) is in agreement with increase in precipitation seasonality reported since the 747 early Badenian. This climatic change was mainly result of tectonically inferred sea-level falls 30 748 in West Carpathian-North Pannonian Basins (Kováč et al., 2001), as well as a possible 749 decrease of moist air mass production during the early Badenian (Böhme, 2003). The second 750 period of increased precipitation seasonality 14.7‒14.5 Ma (Böhme, 2003), which most 751 probably persisted in the northwestern part of the Vienna Basin up to the early/late Badenian 752 transition, was followed by a late Badenian steep decrease of humidity in Central and Eastern 753 Europe. This drop in humidity culminated during the latest Badenian and Sarmatian stages 754 (~13‒12 Ma; Böhme et al., 2008) as also documented by the dramatic succession of dry years 755 with irregular precipitation events recorded from isotopic record of Serravallian oyster shells 756 from the Vienna and Korneuburg Basins (Harzhauser et al., 2011). 757 758 7. Conclusions 759 Here we provide an updated faunal list of the middle Miocene vertebrate assemblage from 760 Czujan’s sandpit site as well as interpretation of the geological, taphonomical and 761 paleoenvironmental data based on the osteological and dentognathic materials. A total of 14 762 vertebrate taxa were indentified, including two reptiles (Testudines: Trionychidae indet. and 763 Testudo kalksburgensis) and 12 mammals (Proboscidea: Prodeinotherim cf. bavaricum, 764 Zygolophodon turicensis, Gomphotherium angustidens; Rhinocerotidae: Hoploaceratherium 765 sp., Brachypotherium brachypus; Chalicotheriidae: Anisodon grande, Equidae: 766 Anchitherium sp.; Suidae: cf. Retroporcus matritensis; Palaeomerycidae indet.; Cervidae: 767 Heteroprox larteti; Bovidae: Tethytragus stehlini and Carnivora: Amphicyon cf. major). The 768 taphonomic analysis indicates that terrestrial vertebrate fossils could not be redeposited. 769 Fossil vertebrates from Czujan’s sandpit come from deposits of a braid delta plain to delta 770 front deposited during the Kosovian substage (late Badenian). Based on the combination of 771 biostratigraphic and geological data, we assume that Czujan’s sandpit age can be placed to 772 latest MN6 for Central Europe (close to Wielician/Kosovain substage boundary: ~13.6 Ma). 31 773 We conclude that there are two plausible explanations for the genesis of the fossil 774 accumulation from Czujan’s sandpit, including: 1) an abiotic time-averaged concentration 775 generated by the transport energy of water; or 2) the gradual or periodical transport of remains 776 from one or more (in the case of more time-averaged as well) mass death site accumulation. 777 However, under both interpretations, the Czujan’s sandpit vertebrate assemblage would 778 reflect an averaged spectrum of species from the whole river basin flowing into northwest of 779 the Vienna Basin. We suggest a mosaic of continental habitats in the area from northwestern 780 coast of the Vienna Basin to the adjacent Carpathian Foredeep Basin. These habitats were 781 represented by the restricted riparian and floodplain forests, surrounded by the widespread 782 woodlands opening locally into more open environment, with enough freshwater sources that 783 could eventually evaporate on a seasonal basis during droughts, leading to mass deaths of 784 animals. 785 786 Authors Contributions 787 J.B. and M.I. conceived and designed the study. J.B. prepared and catalogued the material 788 from MZM, compiled and analysed data from all studied specimens, and prepared the initial 789 manuscript draft (geology, taphonomy, paleoecology, biostratigraphy). J.B, M.I and À.H.L 790 prepared figures and tables. D.M.A., M.I., M.H., and À.H.L. analyzed data and wrote 791 additional sections of the manuscript (M.H.: sedimentology; D.M.A: Suidae and 792 Deinotheriidae; À.H.L.: Testudines and Carnivora). All authors improved the initial 793 manuscript draft equally and approved its final form. 794 795 Acknowledgments 796 32 797 We are indebted to R. Brzobohatý (ÚGV) and M. Bubík (Czech Geological Survey) for the 798 identification of microfossils, M. Roblíčková (MZM), P. Macháček (RMM) and D. Nagel 799 (University of Vienna, Austria) for help and access to collections, and J. Abella (Institut 800 Català de Paleontologia Miquel Crusafont, Spain), R.L. Bernor (Howard University, USA), 801 M.C. Coombs (University of Massachusetts, USA), U.B. Göhlich (Naturhistorisches Museum 802 Wien, Austria), K. Heissig (SNSB - Bayerische Staatssammlung für Paläontologie und 803 Geologie, Germany), M. Pickford (Muséum National d'Histoire Naturelle, France), G. 804 Rössner (SNSB - Bayerische Staatssammlung für Paläontologie und Geologie, Germany), and 805 C. Tóth (Central Slovakia Museum, Slovakia) for discussion on the described fossil material. 806 We would also want to express our gratitude to Martin Sabol (Comenius University in 807 Bratislava, Slovakia) and anonymous reviewer for their constructive reviews that improved 808 our manuscript. This work appears through the institutional support of long-term conceptual 809 development of research institutions provided by the Ministry of Culture of the Czech 810 Republic (ref. MK000094862), the Specific research project at the Faculty of Science at 811 Masaryk University, Brno (MUNI/A/0944/2019), the Spanish Agencia Estatal de 812 Investigación (CGL2016-76431-P, AEI/FEDER EU), the Generalitat de Catalunya (CERCA 813 Programme and 2017 SGR 116), and the Operational Programme Research, Development and 814 Education - Project "Postdoc@MUNI" (No. CZ.02.2.69/0.0/0.0/16_027/0008360). 815 816 References 817 Abel, O., 1910. Erläuterungen zur geologischen Karte der österreichisch-ungarischen 818 Monarchie, NW, Gr., Nr. 85, Zone 10, Koll. XV, Blatt Auspitz und Nikolsburg. Verl. K.- 819 K. Geol. Reichsanst. Wien, 1–37. 33 820 Abusch-Siewert, S., 1983. Gebissmorphologische Untersuchungen an eurasiatischen 821 Anchitherien (Equidae, Mammalia) unter besonderer Berücksichtigung der Fundstelle 822 Sandelzhausen. Cour. Forsch. Senckenberg 62, 1–401. 823 Aiglstorfer, M., Bocherens, H., Böhme, M., 2014. Large mammal ecology in the late Middle 824 Miocene locality Gratkorn (Austria). Paleodivers. Paleoenviron. 94, 189–213. 825 https://doi.org/10.1007/s12549-013-0145-5 826 Alth, A., 1850. Dinotherium giganteum von Nikolsburg. Ber. Mitt. Freunden Naturw. Wien 4, 827 7. 828 Anquetin, J., Antoine, P.-O., Tassy, P., 2007. Middle Miocene Chalicotheriinae (Mammalia, 829 Perissodactyla) from France, with a discussion on chalicotheriine phylogeny. Zool. J. Linn. 830 Soc. 151, 577–508. https://doi.org/10.1111/j.1096-3642.2007.00327.x 831 Argot, C., 2010. Morphofunctional analysis of the postcranium of Amphicyon major 832 (Mammalia, Carnivora, Amphicyonidae) from the Miocene of Sansan (Gers, France) 833 compared to three extant carnivores Ursus arctos, Panthera leo and Canis lupus. 834 Geodiversitas 32, 65–106. https://doi.org/10.5252/g2010n1a2 835 Augustí, J., Antón, M., 2002. Mammoths, Sabertooths, and Hominids: 65 Million Years of 836 Mammalian Evolution in Europe. Columbia University Press, New York, 328 pp. 837 Azanza, B., Morales, J., 1994. Tethytragus nov. gen. et. Gentrytragus nov. gen. Deux 838 nouveaux Bovidés (Artiodactyla, Mammalia) du Miocène moyen. Relations 839 phylogénétiques des Bovidés anté-vallésiens. Proc. K. Ned. Akad. Wet. 97, 249–282. 840 Bachmayer, F., Młynarski, M., 1981. Testudo kalksburgensis Toula, 1896, eine valide 841 Schildkrötenart aus den miozänen Stranbildungen von Kalksburg bei Wein. Sitzungsber. 842 Österr. Akad. Wiss. Math.-Naturwiss. Kl. 190, 111–119. 34 843 Backwell, L., Steininger, C., Neveling, J., Abdala, F., Pereira, L., Mayer, E., Rossouw, L., de 844 la Peña, P., Brink, J., 2018. Holocene large mammal mass death assemblage from South 845 Africa. Quat. Int. 495, 49–63. https://doi.org/10.1016/j.quaint.2017.11.055 846 Becker, D., 2003. Paléoécologie et paléoclimats de la Molasse du Jura (Oligo-Miocène): 847 apport des Rhinocerotoidea (Mammalia) et des minéraux argileux. GeoFocus 9, 1–328. 848 Behrensmeyer, A.K., 1978. Taphonomic and ecologic information from bone weathering. 849 Paleobiol. 4, 150–162. https://doi.org/10.1017/S0094837300005820 850 Behrensmeyer, A.K., 1991. Terrestrial vertebrate accumulations. In: Allison, P.A., Briggs, 851 D.E.G. (Eds.), Taphonomy: Releasing the data locked in the fossil record. Plenum Press, 852 New York, pp. 291–335. 853 Behrensmeyer, A.K., 2007. Bonebeds through Time. In: Rogers, R.R., Eberth, A.D., Fiorillo, 854 R.A. (Eds.), Bonebeds: genesis, analysis, and paleobiological significance. University of 855 Chicago Press, Chicago, pp. 65–97. 856 Bibi, F., Güleç, E.S., 2008. Bovidae (Mammalia: Artiodactyla) from the Late Miocene of 857 Sivas, Turkey. J. Vertebr. Paleontol. 28, 501–519. 858 https://doi.org/10.1671/0272-4634(2008)28[501:BMAFTL]2.0.CO;2 859 Bimka, J., Guryča, I., Heyna, M., Holzknecht, M., Jelínková, Ž., Koukolíček, R., Králová, Z., 860 Procházka, P., Řehánek, J., Šelle, M., Kotásek, V., Vala, Z., Valášek, S., 1983. Závěrečná 861 zpráva o hlubokém strukturním vrtu Nové Mlýny-2. MS Geofond (signature GF P051434), 862 Czech Geological Survey, Prague. 863 Böhme, M., 2003. Miocene Climatic Optimum: evidence from Lower Vertebrates of Central 864 Europe. Palaeogeogr. Palaeoclim. Palaeoecol. 195, 389–401. 865 https://doi.org/10.1016/S0031-0182(03)00367-5 866 Böhme, M., Ilg, A., Winklhofer, M., 2008. Late Miocene “washhouse“ climate in Europe. 867 Earth. Planet. Sc. Lett. 275, 393–401. https://doi.org/10.1016/j.epsl.2008.09.011 35 868 Böhme, M., Winklhofer, M., Ilg A., 2011. Miocene precipitation in Europe: Temporal trends 869 and spatial gradients. Palaeogeogr. Palaeoclim. Palaeoecol. 304, 212–218. 870 https://doi.org/10.1016/j.palaeo.2010.09.028 871 Böhme, M., Aiglstorfer, M., Uhl, D., Kullmer, O., 2012. The antiquity of the Rhine River: 872 stratigraphic coverage of the Dinotheriensande (Eppelsheim Formation) of the Mainz 873 Basin (Germany). PLoS One 7, e36817. https://doi.org/10.1371/journal.pone.0036817 874 Brzobohatý, R., Stráník, Z., 2012. Paleogeography of the Early Badenian connection 875 between the Vienna Basin and the Carpathian Foredeep. Cent. Eur. J. Geosci. 4, 126– 876 137. https://doi.org/10.2478/s13533-011-0045-z 877 Brzobohatý, R., Nolf, D., Kroupa, O., 2007. Fish otoliths from the Middle Miocene of 878 Kienberg at Mikulov, Czech Republic, Vienna Basin: their paleoenvironmental and 879 paleogeographic significance. Bull. Inst. R. Sci. Nat. Belgique Sci. Terre 77, 167–196. 880 Březina, J., 2019. Neogenní savci z okolí Mikulova. Acta Mus. Morav. Sci. Geol. 104, 217– 881 239. 882 Buday, T., Cicha, I., Čtyroký, P., Fejfar, O., 1964. Die Stellung des Neogens der 883 Westkarpaten in der Paratethys. Cursillos Conf. Inst. Lucas Mallada 9, 109–116. 884 Calandra, I., Göhlich, U.B., Merceron, G., 2008. How could sympatric megaherbivores 885 coexist? Example of niche partitioning within a proboscidean community from the 886 Miocene of Europe. Naturwissenschaften 95, 831–838. 887 https://doi.org/10.1007/s00114-008-0391-y 888 Calandra, I., Göhlich, U.B., Merceron, G., 2010. Feeding preferences of Gomphotherium 889 subtapiroideum (Mammalia, Proboscidea) from the Miocene of Sandelzhausen (Northern 890 Alpine Foreland Basin, southern Germany) through life and geological time: Evidence 891 from dental microwear analysis. Paläontol. Z. 84, 205–215. 892 https://doi.org/10.1007/s12542-010-0054-0 36 893 Cerdeño, E., 1993. Étude sur Diaceratherium aurelianense et Brachypotherium brachypus 894 (Rhinocerotidae, Mammalia) du Miocène moyen de France. Bull. Mus. Natl. Hist. Nat. 4, 895 25–77. 896 Cerha, J., 1987. Zpráva o stavebně geologickém průzkumu pro PP Mikulov - Valtická - 897 Zemník. MS, Geofond (signature GF P060807). Czech Geological Survey, Prague 898 Čerňanský, A., Tóth, C., Šurka, J., 2012. Nálezy krokodíla a korytnačiek zo spodného 899 miocénu lokality Baňa Dolina vo Veľkom Krtíši (Slovensko). Acta Geol. Slovaca 4, 113– 900 123. 901 Chen, L., Lu, Y., Wu, J., Xing, F., Liu, L., Ma, Y., Rao, D., Peng, L., 2015. Sedimentary 902 facies and depositional model of shallow water delta dominated by fluvial for Chang 8 oil- 903 bearing group of Yanchang Formation in southwestern Ordos Basin, China. J. Cent. South 904 Univ. 22, 4749–4763. https://doi.org/10.1007/s11771-015-3027-3 905 Costeur, L., Guérin, C., Madriet, O., 2012. Paléoécologie et paléoenvironnement du site 906 miocène de Sansan. In: Peigné, S., Sen, S. (Eds.), Mammifères de Sansan. Mem. Mus. 907 Natl. Hist. Nat. 203, 661–693. 908 Čtyroký, P., 1989. Pannonian and Pontian at the SW margin of the Wiena Basin near 909 Mikulov. Věstn. Ústřed. Úst. Geol. 64, 257–265. 910 Čtyroký, P., 1999. Geologická mapa 1:500 000 moravské části vídeňské pánve. Zpr. Geology. 911 Výzk. Roce 1998, 88–91. 912 Čtyroký, P., Havlíček, P., Stráník, Z., Pálenský, P., 1995. Geologická a přírodovědná mapa 913 CHKO a BR Pálava 1:25 000. Český geologický ústav, Praha. 914 Čtyroký, P., Adámek, J., Brunnerová, Z., Čekan, V., Čtyroká, J., Dudek, A., Eliáš, M., 915 Eliášová, H., havlíček, P., Hrkal, Z., Klečák, J., Kolejka, J., Molčíková, V., Novák, Z., 916 Stráník, Z., 1990. Vysvětlivky k základní geologické mapě ČSSR 1:25 000 34–141 917 Mikulov. Ústřední ústav geologický, Praha. 37 918 Dasgupta, S., Buatois, L.A., Mángano, M.G., 2016. Living on the edge: evaluating the impact 919 of stress factors on animal-sediment interactions in subenvironments of a shelf-margin 920 delta, the Mayaro Formation, Trinidad. J. Sediment. Res. 86, 1034–1066. 921 https://doi.org/10.2110/jsr.2016.47 922 De Bruijn, H., Daams, R., Daxner-Höck, G., Fahlbusch, V., Ginsburg, L., Mein, P., Morales, 923 J., 1992. Report of the RCMNS working group on fossil mammals, Reisensburg 1990. 924 Newsl. Stratigr. 26, 65–118. 925 Delfino, M., Luján, À.H., Carmona, R., Alba, D.M., 2012. Revision of the extinct Pleistocene 926 tortoise Testudo lunellensis Almera and Bofill, 1903 from Cova de Gràcia (Barcelona, 927 Spain). Amphibia-Reptilia 33, 215–225. https://doi.org/10.1163/156853812X636466 928 DeMiguel, D., Azanza, B., Morales, J., 2011. Paleoenvironments and paleoclimate of the 929 Middle Miocene of central Spain: A reconstruction from dental wear of ruminants. 930 Palaeogeogr. Palaeoclim. Palaeoecol. 302, 452–463. 931 https://doi.org/10.1016/j.palaeo.2011.02.005 932 Domingo, L., Cuevas-González, J., Grimes, S.T., Hernández Fernández, M., López-Martínez, 933 N., 2009. Multiproxy reconstruction of the palaeoclimate and palaeoenvironment of the 934 Middle Miocene Somosaguas site (Madrid, Spain) using herbivore dental enamel. 935 Palaeogeogr. Palaeoclim. Palaeoecol. 272, 53–68. 936 https://doi.org/10.1016/j.palaeo.2008.11.006 937 Domingo, L., Koch, P.L., Grimes, S.T., Morales, J., López-Martínez, N., 2012. Isotopic 938 paleoecology of mammals and the Middle Miocene Cooling event in the Madrid Basin 939 (Spain). Palaeogeogr. Palaeoclim. Palaeoecol. 339–341, 98–113. 940 https://doi.org/10.1016/j.palaeo.2012.04.026 941 Eberth, A.D., Rogers, R.R., Fiorillo, R.A., 2007. A practical approach to the study of 942 bonebeds. In: Rogers, R.R., Eberth A.D., Fiorillo, R.A. (Eds.), Bonebeds: Genesis, 38 943 Analysis, and Paleobiological Significance. University of Chicago Press, Chicago, pp. 944 265–332. 945 Estes, R. 1969. Die Fauna der miozäner Spaltefüllung von Neudorf an der March (ČSSR), 946 Reptilia (Lacertilia). Sitz. Österr. Akad. Wiss., Math. Naturwiss. Klasse, 178, 77–82. 947 Fejfar, O., 1990. The Neogene vertebrate paleontology sites of Czechoslovakia: A 948 contribution to the Neogene terrestric biostratigraphy of Europe based on rodents. In: 949 Lindsay, E.H., Fahlbusch, V., Mein, P. (Eds.), European Neogene Mammal Chronology. 950 Plenum Press, New York, pp. 211–236. 951 Francírek, M., 2018. Sedimentárně geologické studium neogenních sedimentů karpatské 952 předhlubně v oblasti Vizovických vrchů. PhD. Thesis, 103 pp., Masaryk University. 953 Gasamans, N., Luján, À.H., Pons-Monjo, G., Obradó, P., Casanovas-Vilar, I., Alba, D.M., in 954 press. The record of Prodeinotherium in the Iberian Peninsula: New data from the Vallès- 955 Penedès Basin. J. Mamm. Evol. https://doi.org/10.1007/s10914-021-09543-y 956 Georgalis, G.L., Joyce, W.G., 2017. A review of the fossil record of Old World turtles of the 957 clade Pan-Trionychidae. Bull. Peabody Mus. Nat. Hist. 58, 115–208. 958 https://doi.org/10.3374/014.058.0106 959 Ginsburg, L., 1961. La Faune des carnivores Miocènes de Sansan. Mem. Mus. Natl. Hist. Nat. 960 9, 1–190. 961 Ginsburg, L., Antunes M.T., 1968. Amphicyon giganteus, Carnassier géant du Miocene. Ann. 962 Paléontol. 54, 1–32. 963 Ginsburg, L., Chevrier, F., 2001. Les Dinothères du bassin de la Loire et l'évolution du genre 964 Deinotherium en France. Symbioses 5, 9–24. 965 Göhlich, U.B., 1998. Elephantoidea (Proboscidea, Mammalia) aus dem Mittel- und 966 Obermiözan der Oberen Süsswassermolasse Süddeutschlands: Odontologie und 967 Osteologie. Münch. Geowiss. Abh. A36, 1–245. 39 968 Göhlich, U.B., 2010. The Proboscidea (Mammalia) from the Miocene of Sandelzhausen 969 (southern Germany). Paläontol. Z. 84, 163–204. 970 https://doi.org/10.1007/s12542-010-0053-1 971 Green, J.L., Semprebon, G., Solounias, N., 2005. Reconstructing the paleodiet of Florida 972 Mammut americanum via low-magnification stereomicroscopy. Palaeogeogr. 973 Palaeoclimatol. Palaeoecol. 223, 34–48. https://doi.org/10.1016/j.palaeo.2005.03.026 974 Guérin C., 2012. Anisodon grande (Perissodactyla, Chalicotheriidae) de Sansan. In: Peigné, 975 S., Sen, S. (Eds.), Mammifères de Sansan. Mem. Mus. Natl. Hist. Nat. 203, 279–315. 976 Harzhauser, M., Piller, W.E., Mülleger, S., Grunert, P., Micheels, A., 2011. Changing 977 seasonality patterns in Central Europe from Miocene Climate Optimum to Miocene 978 Climate Transition deduced from the Crassostrea isotope archive. Glob. Planet. Change 76, 979 77–84. https://doi.org/10.1016/j.gloplacha.2010.12.003 980 Harzhauser, M., Grunert, P., Mandic, O., Lukeneder, P., García Gallardo, Á., Neubauer T.A., 981 Carnevale G., Landau B.M., Sauer R., Strauss P., 2018. Middle and Late Badenian 982 palaeoenvironments in the northern Vienna Basin and their potential link to the Badenian 983 Salinity Crisis. Geol. Carpath. 69, 149–168. https://doi.org/10.1515/geoca-2018-0009 984 Haynes, G., 1985. Age profiles in elephant and mammoth bone assemblages. Quat. Res. 24, 985 333–345. https://doi.org/10.1016/0033-5894(85)90055-9 986 Haynes, G., 1988. Mass deaths and serial predation: Comparative taphonomic studies of 987 modern large mammal death sites. J. Archaeol. Sci. 15, 219–235. 988 https://doi.org/10.1016/0305-4403(88)90064-7 989 Haynes G., 1991. Mammoths, Mastodonts and Elephants. Biology, Behavior and the Fossil 990 Record. Cambridge University Press, Cambridge. 991 Haynes, G., 2017. Finding meaning in mammoth age profiles. Quat. Int. 443A, 65–78. 992 https://doi.org/10.1016/j.quaint.2016.04.012 40 993 Haynes, G., 2018. Raining more than cats and dogs: Looking back at field studies of 994 noncultural animal-bone occurrences. Quat. Int. 466B, 113–130. 995 https://doi.org/10.1016/j.quaint.2015.10.119 996 Haynes, G., Klimowicz, J., 2015. Recent elephant-carcass utilization as a basis for 997 interpreting mammoth exploitation. Quat. Int. 359–360, 19–37. 998 https://doi.org/10.1016/j.quaint.2013.12.040 999 Heckeberg, N.S., 2017. Origination of antlerogenesis. J. Morphol. 278, 182–202. 1000 https://doi.org/10.1002/jmor.20628 1001 Heissig, K., 1999. Family Rhinocerotidae. In: Rössner, G., Heissig, K. (Eds.), The Miocene 1002 Land Mammals of Europe. Dr. Friedrich Pfeil, München, pp. 175–188. 1003 Heissig, K., 2009. The early Vallesian vertebrates of Atzelsdorf (Late Miocene, Austria) 11. 1004 Rhinocerotidae and Chalicotheriidae (Perissodactyla). Ann. Nathist. Mus. Wien 111A, 1005 619–634. 1006 Heissig, K., 2012. Les Rhinocerotidae (Perissodactyla) de Sansan. In: Peigné, S., Sen, S. 1007 (Eds.), Mammifères de Sansan. Mem. Mus. Natl. Hist. Nat. 203, 317–485. 1008 Hilgen, F.J., Lourens, L.J., Van Dam, J.A., Beu, A.G., Boyes, A.F., Cooper, R.A., Krijgsman, 1009 W., Ogg, J.C., Piller, W.E., Wilson, D.S., 2012. The Neogene period. In: Gradstein, F.M., 1010 Ogg, J.G., Schmitz, M.D., Ogg, G.M. (Eds.), The Geologic Time Scale 2012, Volume 2. 1011 Elsevier, Amsterdam, pp. 923–978. 1012 Hohenegger, J., Ćorić, S., Wagreich, M., 2014. Timing of the Middle Miocene Badenian 1013 Stage of the Central Paratethys. Geol. Carpath. 65, 55–66. 1014 https://doi.org/10.2478/geoca-2014-0004 1015 Holec, P., 1985. Finds of mastodon/Proboscidea, Mammalia/relics in Neogene and 1016 Quaternary sediments of Slovakia/ČSSR. Západn. Karp. sér. Paleontol. 10, 13–53. 41 1017 Hörnes, M., 1848. Dinotherium von Nikolsburg. Berichte über die Mitteilungen von Freunden 1018 der Naturwissenschaften in Wien 3, 160–161. 1019 Hutson, J.M., Burke, C.C., Haynes, G., 2013. Osteophagia and bone modifications by giraffe 1020 and other large ungulates. J. Archaeol. Sci. 40, 4139–4149. 1021 https://doi.org/10.1016/j.jas.2013.06.004 1022 Hünermann, K.A., 1989. Die Nashornskelette (Aceratherium incisivum Kaup 1832) aus dem 1023 Jungtertiär vom Höwenegg im Hegau (Südwestdeutschland). Andrias 6, 1–116. 1024 Jiříček, R., 2002. Molasový vývoj alpsko-karpatské předhlubně a vídeňské pánve. Explor. 1025 Geophys. Remote Sens. Environ. 9, 4–178. 1026 Jüttner, K., 1938. Das Neogen des unteren Thayalandes. Verhandlungen Geol. Bundesanst. 1027 32, 201–224. 1028 Jüttner, K., 1939a. Aufschlüsse als geologische Dokumente im unteren Thayalande. Jahrb. 1029 Reichsanst. Bodenforsch. Zweigstelle Wien 89B, 377–411. 1030 Jüttner, K., 1939b. Geologische Karte des Unteren Thayalandes. Mitteilungen der Reichsstelle 1031 für Bodenforschung, 1, 1. Wien. 1032 Jüttner, K., 1940. Erläuterungen zur geologischen Karte des unteren Thayalandes. Mitt. 1033 Reichsstelle Bodenforsch. 1, 1–57. 1034 Kaiser, T., 2009. Anchitherium aurelianense (Equidae, Mammalia): a brachydont ―dirty 1035 browser in the community of herbivorous large mammals from Sandelzhausen (Miocene, 1036 Germany). Paläontol. Z. 83, 131–140. https://doi.org/10.1007/s12542-009-0002-z 1037 Kázmér, M., 2008. The Miocene Bükkábrány Fossil Forest in Hungary – field observations 1038 and project outline. Hantkeniana 6, 229–244. 1039 Koufos, G.D., 2016. Neogene and Quarternary continental biostratigraphy of grece based on 1040 mammals. Bull. Geol. Soc. Greece 50, 55–64. https://doi.org/10.12681/bgsg.11701 42 1041 Kováč, M., Nagymarosy, A., Holcová, K., Hudáčková, N., Zlinská, A., 2001. 1042 Paleogeography, paleoecology and eustasy: Miocene 3rd order cycles of relative sea-level 1043 changes in the Western Carpathian-North Pannonian basins. Acta Geol. Hung. 44, 1–45. 1044 Kováč, M., Andreyeva-Grigorovich, A., Bajraktarević, Z., Brzobohatý, R., Filipescu, S., 1045 Fodor, L., Harzhauser, M., Nagymarosy, A., Oszczypko, N., Pavelić, D., Rögl, F., Saftić, 1046 B., Sliva, L., Studencka, B., 2007. Badenian evolution of the Central Paratethys Sea: 1047 paleogeography, climate and eustatic sea-level changes. Geol. Carpath. 58, 579–606. 1048 Kováč, M., Halásová, E., Hudáčková, N., Holcová, K., Hyžný, M., Jamrich, M., Ruman, A., 1049 2018. Towards better correlation of the Central Paratethys regional time scale with the 1050 standard geological time scale of the Miocene Epoch. Geol. Carpath. 69, 3, 283–300. 1051 https://doi.org/10.1515/geoca-2018-0017 1052 Kováčová, M., Doláková N., Kováč, M., 2011. Miocene vegetation pattern and climate 1053 change in the northwestern Central Paratethys domain (Czech and Slovak Republic). Geol. 1054 Carpath. 62, 251–266. https://doi.org/10.2478/v10096-011-0020-0 1055 Kovar-Eder, J., Teodoridis, V., 2018. The Middle Miocene Central European plant record 1056 revisited; widespread subhumid sclerophyllous forests indicated. Fossil Imprint, 74, 115– 1057 134. https://doi.org/10.2478/if-2018-0009 1058 Kuklová, K., 1970. Zpráva o provedení doplňujícího hydrogeologického průzkumu. 1059 Pohořelice II–III. etapa. Lokalita Mikulov. MS, Geofond (signature GF P022573). Czech 1060 Geological Survey, Prague. 1061 Kuklová, K., 1978. Závěrečná zpráva o regionálním hydrogeologickém průzkumu neogenních 1062 uloženin jihozápadní části rajónu R 33. MS, Geofond (signature GF FZ 005709). Czech 1063 Geological Survey, Prague. 43 1064 Kuss, S., 1965. Revision der europäischen Amphicyoninae (Canidae, Carnivora, Mammalia) 1065 ausschliesslich der voroberstampischen Formen. Sitzungsberichte der Heidelberg 1066 Akademie der Wissenschaften, Mathematisch-naturwissenschaftliche Klasse 1, 1–168. 1067 Kvaček, Z., Böhme, M., Dvořák, Z., Konzalová, M., Mach, K., Prokop, J., Rajchl, M., 2004. 1068 Early Miocene freshwater and swamp ecosystems of the Most Basin (north Bohemia) with 1069 particular reference to the Bilina Mine section. J. Czech Geol. Soc. 49, 1–40. 1070 Kvaček, Z., Kováč, M., Kovar-Eder, J., Doláková, N., Jechorek, H., Parashiv, V., Kováčová, 1071 M., Sliva, L., 2006. Miocene evolution of landscape and vegetation in the Central 1072 Paratethys. Geol. Carpath. 57, 295–310. 1073 Lambert, W.D., 1996. The biogeography of the gomphotheriid proboscideans of North 1074 America. In: Shoshani, J., Tassy, P., (Eds.), The Proboscidea: Evolution and Palaeoecology 1075 of Elephants and Their Relatives. Oxford University Press, Oxford, pp. 143–148. 1076 Lambert, W.D., Shoshani, J., 1998. Proboscidea. In: Janis, C.M., Scott, K.M., Jacobs, L.L. 1077 (Eds.), Evolution of Tertiary Mammals of North America Volume 1: Terrestrial 1078 Carnivores, Ungulates, and Ungulate-like Mammals. Cambridge University Press, 1079 Cambridge, pp. 606–621. 1080 Lapparent de Broin F., 2000. Les Chéloniens de Sansan. In: Ginsburg, L., (Ed.), La faune 1081 Miocéne de Sansan et son environment. Mem. Mus. Nat. Hist. Nat. 183, 219–261. 1082 Luján, À.H., 2015. Les tortugues (Testudines) del Neogeni Quaternari de Catalunya (NE de la 1083 Península Ibèrica). PhD. Thesis. Universitat Autònoma de Barcelona, Barcelona, pp. 282. 1084 Luján, À.H., Delfino, M., Casanovas-Vilar, I. Alba, D.M., 2014. Taxonomy of subgenus 1085 Temnoclemmys Bergounioux, 1958 (Testudines: Geoemydidae: Ptychogasterinae) based on 1086 new material from the Vallès-Penedès Basin. C. R. Palevol 13, 277–295. 1087 https://doi.org/10.1016/j.crpv.2013.11.004 44 1088 Luján, À.H., Delfino, M., Robles, J.M., Alba, D.M., 2016. A review of the Miocene tortoise 1089 Testudo catalaunica Bataller, 1926 and a revised phylogeny of extinct species of genus 1090 Testudo (Testudines: Testudinidae). Zool. J. Linn. Soc. 178, 312–342. 1091 https://doi.org/10.1111/zoj.12414 1092 Luján, À.H., Čerňanský, A., Bonilla-Salomón, I., Březina, J., Ivanov M., in press. Fossil 1093 turtles from the early Miocene localities of Mokrá-Quarry (Burdigalian, MN4), South 1094 Moravian Region, Czechia. Geodiversitas. 1095 Marmi, J., Luján, À.H., 2012. An overview of the threatened phylogenetic diversity of living 1096 testudines based on a review of the complex evolutionary history of turtles. In: Cosgrove, 1097 M.J., Roe, S.A., (Eds.), Turtles: Anatomy, Ecology and Conservation. Nova Science 1098 Publishers, New York, pp. 117–150. 1099 Mein, P., 1975. Résultats du groupe de travail des vertébrés: Biozonation du Néogène 1100 méditerranéen à partir des mammifères. In: Senes, J. (Ed.), Report on Activity of the 1101 RCMNS Working Groups (1971–1975). Slovak Academy of Sciences, Bratislava, pp. 78– 1102 81. 1103 Mein, P., 1990. Updating of MN zones. In: Lindsay, E.H., Fahlbusch, V., Mein, P. (Eds.), 1104 European Neogene Mammal Chronology. Plenum Press, New York, pp. 73–90. 1105 Mein, P., 1999. European Miocene Mammal Biochronology. In: Rössner, G.E., Heissig, K. 1106 (Eds.), The Miocene Land Mammals of Europe. Dr. Friedrich Pfeil, München, pp. 25–38. 1107 McPherson, J.G., Shanmugam, G., Moiola, R.J., 1988. Fan deltas and braid deltas: conceptual 1108 problems. In: Nemec, W., Steel, R.J. (Eds.), Fan Deltas: Sedimentology and Tectonic 1109 Settings. Blackie, Glasgow, pp. 14–22. 1110 Miall, A.D., 1996. The geology of fluvial deposits: sedimentary facies, basin analysis and 1111 petroleum Geology. Springer-Verlag, Berlin, 582 pp. 45 1112 Miklas-Tempfer, P.M., 2003. The Miocene Herpetofaunas of Grund (Caudata; Chelonii, 1113 Sauria, Serpentes) and Mühlbach am Manhartsberg (Chelonii, Sauria, Amphisbaenia, 1114 Serpentes), Lower Austria. Ann. Naturhist. Mus. Wien. 104A, 195–235. 1115 Miklas-Tempfer, P.M., 2005. The Herpetofauna (Amphibia: Caudata, Anura; Reptilia: 1116 Sceroglossa) of the Upper Miocene Locality Kohfidisch (Burgenland, Austria). Beitr. 1117 Paläontol. 29, 145–253. 1118 Musil, R., 1956. Nález lebky Deinotheria na jižní Moravě. Acta Mus. Morav. Sci. Natural. 41, 1119 65–86. 1120 Nemec, W., 1990. Deltas - remarks on terminology and classification. In: Colella, A., Prior, 1121 D.B. (Eds.), Coarse-grained Deltas. Special Publication of the International Association of 1122 Sedimentologists. Blackwell, Oxford, pp. 3–12. 1123 Olariu, C., Bhattacharya, J.P., 2006. Terminal distributary channels and delta front 1124 architecture of river-dominated delta systems. J. Sediment. Res. 76, 212–233. 1125 https://doi.org/10.2110/jsr.2006.026 1126 Patočková, V., 1966. Užití vybraných základních statistických metod pro běžnou 1127 paleontologickou praxi. M.Sc. Thesis, Masaryk University, 149 pp. 1128 Pickford, M., 2013. Conohyus simorrensis (Lartet, 1851) (Suidae, Mammalia) from the 1129 Middle Miocene of Carpetana (Madrid, Spain). Span. J. Palaeontol. 28, 91–102. 1130 https://doi.org/10.7203/sjp.28.1.17834 1131 Pickford, M., 2016. Biochronology of European Miocene Tetraconodontinae (Suidae, 1132 Artiodactyla, Mammalia) flowing from recent revision of the Subfamily. Ann. Naturhist. 1133 Mus. Wien A 118, 175–244. 1134 Pickford, M., Laurent, Y., 2014. Valorisation of palaeontological collections: nomination of a 1135 lectotype for Conohyus simorrensis (Lartet, 1851), Villefranche d’Astarac, France, and 46 1136 description of a new genus of tetraconodont. Estud. Geol. 70, 1–28. 1137 http://dx.doi.org/10.3989/egeol.41261.262 1138 Pickford, M., Pourabrishami, Z., 2013. Deciphering Dinotheriensande deinotheriid diversity. 1139 Palaeobiodivers. Palaeoenvir. 93, 121–150. https://doi.org/10.1007/s12549-013-0115-y 1140 Postma, G., 1990. Depositional architecture and facies of river and fan deltas: a synthesis. In: 1141 Colella, A., Prior, D.B. (Eds.), Coarse-grained Deltas. Special Publication of the 1142 International Association of Sedimentologists, Blackwell, Oxford, pp. 13–27. 1143 Rogers, R.R., Kidwell, M.S., 2007. A conceptual framework for the genesis and analysis of 1144 vertebrate skeletal concentrations, in: Rogers, R.R., Eberth, A.D., Fiorillo, R.A. (Eds.), 1145 Bonebeds: genesis, analysis, and paleobiological significance. University of Chicago Press, 1146 Chicago, pp. 1–63. 1147 Rotgers, C., Alba, D.M., Robles, J.M., Casanovas-Vilar, I., Galindo, J., Bertó, J.V., Moyà- 1148 Solà, S., 2011. A new species of Anchitherium (Equidae: Anchitheriinae) from the Middle 1149 Miocene of Abocador de Can Mata (Vallès-Penedès Basin, NE Iberian Peninsula). C. R. 1150 Palevol 10, 567–576. https://doi.org/10.1016/j.crpv.2011.06.003 1151 Rössner, G.E., 2010. Systematics and palaeoecology of Ruminantia (Artiodactyla, Mammalia) 1152 from the Miocene of Sandelzhausen (southern Germany, Northern Alpine Foreland Basin). 1153 Paläontol. Z. 84, 123–162. https://doi.org/10.1007/s12542-010-0052-2 1154 Rössner, G.E, Costeur, L., Scheyer, T.M., 2021. Antiquity and fundamental processes of the 1155 antler cycle in Cervidae (Mammalia). Sci. Nat. 108, 1–24. https://doi.org/10.1007/s00114- 1156 020-01713-x 1157 Rzehak, A., 1922. Das Mährische Tertiar. vol. 3. Státní Geologický Ústav Československé 1158 Republiky. Prague, 51 pp. 1159 Sabol, M., Holec, P., 2002. Temporal and spatial distribution of Miocene mammals in the 1160 Western Carpathians. Geol. Carpath. 53, 269–279. 47 1161 Sabol, M., Kováč, M., 2006. Badenian Palaeoenvironment, Faunal Succession and 1162 Biostratigraphy: A Case Study from Northern Vienna Basin, Devínska Nova Ves-Bonanza 1163 site (Western Carpathians, Slovakia). Beitr. Palaont. 30, 415–425. 1164 Schulz, E., Fahlke, J.M., Merceron, G., Kaiser, T.M., 2007. Feeding ecology of the 1165 Chalicotheriidae (Mammalia, Perissodactyla, Ancylopoda). Results from dental micro- and 1166 mesowear analyses. Verhandlung. Naturwiss. Ver. Hambg. 43, 5–32. 1167 Seitl, L., 1985. Fauna mastodontů (Proboscidea, Mammalia) moravské části vídeňské pánve. 1168 Čas. Morav. Muz. 70, 71–94. 1169 Siliceo, G., Morales, J., Antón, M., Salesa, M.J., 2020. New fossils of Amphicyonidae 1170 (Carnivora) from the middle Miocene (MN6) site of Carpetana (Madrid, Spain). 1171 Geodiversitas 42, 223–238. https://doi.org/10.5252/geodiversitas2020v42a15 1172 Steininger, F.F., 1999. Chronostratigraphy, Geochronology and Biochronology of the 1173 Miocene ‘European Land Mammal Mega-Zones’ (ELMMZ) and the Miocene ‘Mammal- 1174 Zones (MN-Zones)’. In: Rössner, G.E., Heissig, K. (Eds.), The Miocene Land Mammals of 1175 Europe. Dr. Friedrich Pfeil, München, pp. 9–24. 1176 Stejskal, J., 1934. Geologická stavba Pavlovských vrchů na jižní Moravě. I. Stratigrafie. 1177 Věstn. Státního Geolo. Ústavu Českoslov. Repub. 10, 199–209. 1178 Stráník Z., Čtyroký, P., Havlíček, P., 1999. Geologická minulost Pavlovských vrchů. Sborník 1179 Geol. Věd. Geol. 49, 5–32. 1180 Tassy, P., 1977. Découverte de Zygolophodon turicensis (Schinz) (Proboscidea, Mammalia) 1181 au lieu-dit Malartic à Simorre, Gers (Vindobonien moyen); implications paléoécologiques 1182 et biostratigraphiques. Geobios 10, 655−669. 1183 https://doi.org/10.1016/S0016-6995(77)80045-4 1184 Tassy, P., 1996. Growth and sexual dimorphism among Miocene elephantoids: the example of 1185 Gomphotherium angustidens. In: Shoshani, J., Tassy, P., (Eds.), The Proboscidea: 48 1186 Evolution and Palaeoecology of Elephants and Their Relatives. Oxford University Press, 1187 Oxford, pp. 92–100. 1188 Tassy, P., 2013. L’anatomie cranio-mandibulaire de Gomphotherium angustidens (Cuvier, 1189 1817) (Proboscidea, Mammalia): données issues du gisement d’En Péjouan (Miocène 1190 moyen du Gers, France). Geodiversitas 35, 377–445. https://doi.org/10.5252/g2013n2a6 1191 Tassy, P., 2014. L’odontologie de Gomphotherium angustidens (Cuvier, 1817) d’En Péjouan 1192 (Miocène moyen du Gers, France). Geodiversitas 36, 35–115. 1193 https://doi.org/10.5252/g2014n1a2 1194 Thenius, E. 1948. Zur Kenntnis der fossilen Hirsche des Wiener Beckens, unter besonderer 1195 Berücksichtigung ihrer stratigraphischen Bedeutung. Ann. Nat. Hist. Mus. Wien. 56, 262– 1196 308. 1197 Thenius, E., 1951. Gazella cf. deperdita aus dem mitteleuropaischen Vindobonien und das 1198 Auftreten der Hipparionfauna. Eclogae Geol. Helv. 44, 381–394. 1199 Thenius, E., 1959. Tertiär. II: Wirbeltierfaunen. Handbuch der Stratigraphischen Geologie. 1200 Ferdinand Enke Verlag, Stuttgart. 1201 Thenius, E., 1979. Zur systematischen Stellung und Verbreitung von “Gazella” stehlini 1202 (Bovidae, Mammalia) aus dem Miozän Europas. Anz. Österr. Akad. Wiss. Math.- 1203 Naturwiss. Kl. 116, 9–13. 1204 Tobien, H., 1973. On the evolution of mastodonts (Proboscidea, Mammalia). Part 1: The 1205 bunodont trilophodont groups. Hess. Landesamt Bodenforsch. 101, 202–276. 1206 Tobien, H., 1975. The structure of the mastodont molar (Proboscidea, Mammalia). Part 2: The 1207 zygodont and zygobunodont patterns. Mainz. Geowiss. Mitt. 4, 195–233. 1208 Tobien H., 1996: Evolution of zygodons with emphasis on dentition. In: Shoshani, J., Tassy, 1209 P. (Eds.), The Proboscidea. Evolution and Paleoecology of Elephants and Their Relatives. 1210 Oxford University Press, Oxford, pp. 335–348. 49 1211 Toula, F., 1896. Ueber neue Wirbelthierreste aus dem Tertiär Oesterreichs und Rumeliens. Z. 1212 Deutsch. Geol. Ges. 48, 915–924. 1213 Tütken, T, Vennemann, T.W., Janz, H., Heizmann, E.P.J., 2006. Palaeoenvironment and 1214 palaeoclimate of the Middle Miocene lake in the Steinheim basin, SW Germany: A 1215 reconstruction from C, O, and Sr isotopes of fossil remains. Palaeogeogr. Palaeoclim. 1216 Palaeoecol. 241, 457–491. https://doi.org/10.1016/j.palaeo.2006.04.007 1217 Tütken, T., Vennemann, T., 2009. Stable isotope ecology of Miocene large mammals from 1218 Sandelzhausen, southern Germany. Paläontol. Z. 83, 207–226. 1219 https://doi.org/10.1007/s12542-009-0011-y 1220 van der Made, J., 1989. A Conohyus-lineage (Suidae, Artiodactyla) from the Miocene of 1221 Europe. Rev. Esp. Paleontol. 4, 19–28. 1222 van der Made, J., 2020. The Suoidea from the Middle Miocene of Gračanica (Bugojno Basin, 1223 Bosnia and Herzegovina)—evolution, taxononomy, and biostratigraphy. Palaeobiodivers. 1224 Palaeoenviron. 100, 321–349. https://doi.org/10.1007/s12549-020-00420-9 1225 van der Made, J., Salesa, M. J., 2004. Early remains of the pig Conohyus simorrensis from the 1226 Middle Aragonian of Somosaguas near Madrid - Its dispersal into Europe and evolution. 1227 N. Jb. Geol. Paläont. Abh. 233, 153–168. https://doi.org/10.1127/njgpa/233/2004/153 1228 Viranta, S., 1996. European Miocene Amphicyonidae: taxonomy, systematics and ecology. 1229 Acta Zool. Fenn. 204, 1–61. 1230 Vitek, N.S., Joyce, W.G., 2015. A review of the fossil record of New World turtles of the 1231 clade Pan-Trionychidae. Bull. Peabody Mus. Nat. Hist. 56, 185–244. 1232 https://doi.org/10.3374/014.058.0106 1233 Zapfe, H., 1949. Eine mittelmiozäne Säugetierfauna aus einer Spaltenfüllung bei Neudorf an 1234 der March (CSR). Sitz. Österr. Akad. Wiss., Math. Naturwiss. Klasse. 173–181. 50 1235 Zapfe, H., 1954. Die Fauna der miozänen Spaltenfüllung von Neudorf a. d. March (CSR.) – 1236 Proboscidea. Sitz. Österr. Akad. Wiss., Math. Naturwiss. Klasse, 163, 71–87. 1237 Zapfe, H., 1979. Chalicotherium grande (Blainv.) aus der miozänen Spaltenfüllung von 1238 Neudorf an der March (Děvinská Nová Ves), Tschechoslowakei. N. Denkschr. Naturhist. 1239 Mus. Wien 2, 1–284. 1240 Zapfe, H., 1993. Die Fauna der miozänen Spaltenfüllung von Neudorf an der March 1241 (Slowakei). Palaeomerycidae. Sitz. Österr. Akad. Wiss., Math. Naturwiss. Klasse. 200, 1– 1242 10, 89–136. 1243 Zhang, L., Bao, Z., Lin, Y., Chen, Y., Lin, X., Dou, L., Kong, B., 2017. Genetic types and 1244 sedimentary model of sandbodies in a shallow-water delta: A case study of the first 1245 Member of Cretaceous Yaojia Formation in Qian’an area, south of Songliao Basin, NE 1246 China. Pet. Explor. Dev. 44, 770–779. https://doi.org/10.1016/S1876-3804(17)30087-3 1247 Zhu, X., Li, S., Wu, D., Zhu, S., Dong, Y., Zhao, D., Wang, X., Zhang, Q., 2017. Sedimentary 1248 characteristics of shallow-water braided delta of the Jurassic, Junggar basin, Western 1249 China. Journal of Petroleum Science and Engineering 149, 591–602. 1250 https://doi.org/10.1016/j.petrol.2016.10.054 51 1251 Figure captions 1252 1253 Fig. 1. A: Simplified geological map of Central Europe (A: Austria, CZ: Czech Republic, G: 1254 Germany, H: Hungary, PL: Poland, SK: Slovakia). B: Geological map of Mikulov area where 1255 the location of Czujan’s sandpit is denoted by an asterisk (based on Jüttner, 1939b; Čtyroký et 1256 al., 1995). 1257 1258 Fig. 2. Geological cross sections through the area of Czujan’s sandpit reconstructed from drill 1259 cores and boreholes: HV1 and HJ1 (Kuklová, 1970, 1978); PV1 and PV3 (Seitl, 1985); S164- 1260 180 (Cerha, 1987). The red line indicates the border between successive upper Badenian 1261 substages Wielician and Kosovian according to the borehole NM2 (Bimka et al., 1983). 1262 1263 Fig. 3. Composite stratigraphic log of the Hrušky Formation in the studied area, including the 1264 fossil content recovered from the borehole samples (Kuklová, 1970, 1978; Seitl, 1985; Cerha, 1265 1987) and stratigraphic boundary between successive upper Badenian substages Wielician 1266 and Kosovian in Nové Mlýny 2 borehole (Bimka et al., 1983). 1267 1268 Fig. 4. Photographs taken by R. Musil in 1964 showing the uppermost part of the Hrušky 1269 Formation. Facies association 3 (FA3, Kosovian, modified with lines indicating the 1270 sedimentary architecture). A: sequence of two sedimentary units (St, SGt and Sh, SGh). B: 1271 sequence of two sedimentary units (Sp facies and Sl, Gl). The irregularity of the originally 1272 horizontally placed clay body is caused by a partial slide of the wall. Original in photoarchive 1273 of the Department of Geology and Paleontology, MZM; colored lines are explained in the 1274 text. 1275 52 1276 Fig. 5. Vertebrate remains from Czujan’s sandpit. A: Trionychidae indet., costal plate (PIUW- 1277 8X1939) in dorsal view. B–C: Testudo (Chersine) kalksburgensis, carapace and plastron 1278 (Ot7875) in dorsal (B) and visceral (C) views. D: Prodeinotherium cf. bavaricum, left M3 1279 (Ot7526), in oclusal view, E: Zygolophodon turicensis, right M3 (Ot7541), in oclusal view. 1280 F–G: Gomphotherium angustidens, right M2 (Ot7524) in (F) oclusal view; right I2 (Ot7505) 1281 in (G) lateral view. H: Anisodon grande, left fourth metacarpal (Ot7766) in lateral view. I: 1282 Hoploaceratherium sp., right tibia (Ot7749) in dorsal view. J–L: Brachypotherium brachypus, 1283 left tibia (Ot7752) in (J) dorsal view; right incisor (Ot7700) in (K) labial view; left maxillary 1284 fragment with M2–M3 (Ot7752) in (L) oclusal view. M: Anchitherium sp., right maxillary 1285 fragment with M1–M2 (Ot7765) in oclusal view. N–O: Tethytragus stehlini, frontoparietal 1286 with horn cores (Ot7786) in anterior (N) and left lateral (O) views. P: Palaeomerycidae indet., 1287 right mandibular fragment with m2–m3 (Ot7773) in oclusal view. Q: Heteroprox larteti, 1288 partial antler (Ot7771) in side view. R: cf. Retroporcus matritensis, right (with m3) and left 1289 (with m1) mandibular fragments (RMM P11) in oclusal view. S–T: Amphicyon cf. major, 1290 partial left canine (Ot7814) in (S) lingual view; left calcaneus (Ot7815) in (T) proximal view. 1291 1292 Fig. 6. Taphonomic features of fossil vertebrate remains from Czujan’s sandpit. A–B: 1293 Proboscidean right (Ot7596) and left (Ot7599) humerus from a single individual in (A–B) 1294 cranial views. C: Proboscidea indet., atlas (Ot7545) and axis (Ot7547) in anatomical 1295 articulation, in dorsal view. D: Hoploaceratherium sp., radius (Ot7738) and ulna (Ot7829) in 1296 anatomical articulation, in distal view. E: Artiodactyla indet. (small size), tibia (Ot7810) and 1297 astragalus (Ot7808) in anatomical articulation, in plantar view. F: Proboscidea indet., right 1298 femur (Ot7647) with unfused distal epiphysis (Ot7650), in medial view. G: Right posterior 1299 carapace fragment with well preserved growth scute lines (Ot7877) in lateral view. H–I: 1300 Rhinocerotidae indet., comparison of a well preserved radius (Ot7736) with a deformed 53 1301 partial radius (Ot7737) as a possible result of trampling, in (H–I) caudal views. J: 1302 Brachypotherium brachypus, cracks in distal head of humerus (Ot7720), in caudal view. K–L: 1303 Proboscidea indet., proximal head of humerus (Ot7646) with gnawing traces of a large 1304 carnivore, in proximal (K) and caudal (L) views. M: Proboscidea indet., detail of gnawing 1305 traces of on the tip of tusk (Ot7483) possibly produced by a small rodent. N: Isurus sp., partial 1306 tooth (Ot7870), in labial view. O: wood fragment (Ot7871). 1307 1308 Fig 7. Stratigraphic distribution of the studied taxa in Central Europe. Age boundaries for the 1309 Badenian stage follow Kováč et al. (2018). Modified MN zonation in Western and 1310 Southwestern Europe follows Hilgen et al. (2012), Central European MN zonation modified 1311 according to Steininger (1999), Becker (2003) and Hilgen et al. (2012). The taxa ranges and 1312 ages of localities are based on various sources (see Supplementary Information). 1313 1314 Fig. 8. Paleoenvironments derived for Czujan’s sandpit taxa. Taxa silhouettes are not in scale: 1315 1) Zygolophodon turicensis; 2) Heteroprox larteti; 3) Anisodon grande; 4) Palaeomerycidae; 1316 5) Tethytragus stehlini ; 6) Amphicyon major; 7) Prodeinotherium bavaricum; 8) 1317 Brachypotherium brachypus; 9) Gomphotherium angustidens; 10) Hoploaceratherium; 11) 1318 Anchitherium sp.; 12) Retroporcus matritensis; 13) Testudo kalksburgensis. 1319 1320 Table captions 1321 Table 1 Comparison of minimum numbers of individuals (MNI) recorded at Czujan’s sandpit 1322 with recent mass death sites reported during drought in Zimbabwe (Haynes, 1988, counted in 1323 1986). 1324 54 1325 Table 2 Age profile of the studied proboscideans (Zygolophodon turicensis and 1326 Gomphotherium angustidens) from Czujan’s sandpit. Dental ages follow Tassy (1996, 2013, 1327 2014), whereas ontogenetic age and age categories follow Haynes (2017). 1328 1329 Supplementary material 1330 SUPPLEMENTARY TABLE 1 1331 List of studied material with distribution of skeletal elements for each determined taxa. 1332 1333 SUPPLEMENTARY INFORMATION. 1334 List of well dated Neogene localities in Central Europe with the occurrence of the taxa 1335 recorded at Czujan’s sandpit. Countries are given in capital letters and the name of the locality 1336 is in bold. 1337 55 Table 1 Comparison of minimum numbers of individuals (MNI) recorded at Czujan’s sandpit with recent mass death sites reported during drought in Zimbabwe (Haynes, 1988, counted in 1986). Taxon Czujan’s sandpit Shabi Shabi Lememba Shkawanki Nehimba Proboscidea 21 45 21 8 4 Non-equid Perissodactyla 7 - - - - Equidae 1 2 2 - - Non-bovid Ruminantia 4 - 3 1 - Bovidae 11 11 8 1 2 Suidae 1 - - 1 1 Carnivora 1 2 2 - - Aves - 5 1 - - Testudines 4 4 - - 1 Total MNIs 50 69 37 11 8 Taxa represented 14 13 8 4 5 Table 2 Age profile of studied proboscideans (Zygolophodon turicensis and Gomphotherium angustidens) from Czujan’s sandpit. Dental ages follow Tassy (1996, 2013, 2014), whereas ontogenetic age follow Haynes (2017). MNI Approximate age Dental ages Zygolophodon turicensis Gomphotherium angustidens (years) I-II. 0‒1 III. 1 1‒1.5 IV-VI. 1.5‒5.5 VII. 1 5.5‒7 VIII-IX. 7‒10 X. 1 10‒13 XI-XII. 13‒18 XIII. 1 18‒20 XIV. 3 1 20‒22 XV. 2 22‒24 XVI. 4 24‒27 XVII. 1 27‒28 XVIII-XIX. 28‒32 XX. 2 1 32‒35 XXI. 35‒37 XXII. 1 37‒42 1 SUPPLEMENTARY INFORMATION 2 List of well dated Neogene localities in Central Europe with the occurrence of the taxa 3 recorded at Czujan’s sandpit. Countries are given in capital letters and the name of the locality 4 is in bold. 5 6 AUSTRIA 7 Grund 8 Age: ⁓15.2 Ma (Daxner-Höck et al., 2004) 9 Taxa: Locality Supraspecific taxa Species References Grund Proboscidea Prodeinotherium bavaricum Daxner-Höck et al. (2004) Grund Palaeomerycidae Palaeomeryx eminens Daxner-Höck et al. (2004) 10 11 Guntersdorf 12 Age: ⁓15.2 Ma (Daxner-Höck et al., 2004) 13 Taxa: Locality Supraspecific taxa Species References Guntersdorf Proboscidea Gomphotherium angustidens Daxner-Höck et al. (2004) 14 15 Seegraben near Leoben 16 Age: 14.8 Ma (Aiglstorfer et al., 2014a) 17 Taxa: Locality Supraspecific taxa Species References Seegraben Proboscidea Prodeinotherium bavaricum Aiglstorfer et al. (2014a) Seegraben Cervidae Heteroprox larteti Aiglstorfer et al. (2014a) 18 19 Göriach 20 Age: ⁓14.5 Ma (Aiglstorfer et al., 2014a) 21 1 22 Taxa: Locality Supraspecific taxa Species References Hoploaceratherium Göriach Rhinocerotidae Heissig (2012) tetradactylum Göriach Palaeomerycidae Ampelomeryx magnus Astibia (2012) Pickford and Laurent Göriach Suidae Retroporcus matritensis (2014) 23 24 Klein-hadersdorf 25 Age: ⁓14.2 Ma (Böhme et al., 2012a) 26 Taxa: Locality Supraspecific taxa Species References Klein-hadersdorf Proboscidea Gomphotherium angustidens Thenius (1948) Klein-hadersdorf Proboscidea Zygolophodon turicensis Thenius (1948) Hoploaceratherium cf. Klein-hadersdorf Rhinocerotidae Thenius (1948) tetradactylum Klein-hadersdorf Rhinocerotidae Brachypotherium brachypus Thenius (1948) Klein-hadersdorf Palaeomerycidae Palaeomeryx eminens Thenius (1948) Klein-hadersdorf Cervidae Heteroprox larteti Thenius (1948) Klein-hadersdorf Bovidae Tethytragus stehlini Thenius (1951) 27 28 Gratkorn 29 Age: 12.2–12.0 Ma (Gross et al., 2014) 30 Taxa: Locality Supraspecific taxa Species References Gratkorn Rhinocerotidae Brachypotherium brachypus Gross et al. (2014) Gratkorn Palaeomerycidae Palaeomerycidae indet. Aiglstorfer et al. (2014a) 31 32 SLOVAKIA 33 Devínská Nová Ves–Zapfe’s fissures 34 Age: lower part of MN6, 14.5‒13.82 (Fejfar, 1990) 35 Taxa: Site Supraspecific taxa Species References D.N.V. –Zapfe’s Proboscidea Prodeinotherium bavaricum Sabol and Holec (2002) fissures D.N.V. –Zapfe’s Proboscidea Zygolophodon turicensis Sabol and Holec (2002) 2 fissures D.N.V. –Zapfe’s Hoploaceratherium cf. Rhinocerotidae Sabol and Holec (2002) fissures tetradactylum D.N.V. –Zapfe’s Chalicotheriidae Anisodon grande Sabol and Holec (2002) fissures D.N.V. –Zapfe’s Cervidae Heteroprox larteti Sabol and Holec (2002) fissures D.N.V. –Zapfe’s Palaeomerycidae Palaeomeryx magnus Sabol and Holec (2002) fissures D.N.V. –Zapfe’s Carnivora Amphicyon major Sabol and Holec (2002) fissures 36 37 Devínská Nová Ves–Bonanza 38 Age: late Badenian, 13.82‒13.60 Ma (Sabol and Kováč 2006; Ivanov, in press) 39 Taxa: Site Supraspecific taxa Species References D.N.V. – Proboscidea Zygolophodon turicensis Sabol and Kováč (2006) Bonanza 40 41 Devínská Nová Ves–Sandberg 42 Age: late Badenian, 13.60–13.1 Ma (Hyžný et al., 2012) 43 Taxa: Site Supraspecific taxa Species References D.N.V. – Proboscidea Zygolophodon turicensis Sabol and Holec (2002) Sandberg D.N.V. – Equidae Anchitherium aurelianense Sabol and Holec (2002) Sandberg D.N.V. – Palaeomerycidae Palaeomeryx kaupi Sabol and Holec (2002) Sandberg D.N.V. – Cervidae Heteroprox larteti Sabol and Holec (2002) Sandberg 44 45 SWITZERLAND 46 Hüllistein 47 Age: 15.9–16.0 Ma (Aiglstorfer et al., 2014b) 48 49 Taxa: 3 Locality Supraspecific taxa Species References Hüllistein Proboscidea Gomphotherium angustidens Bollinger (1992) Hüllistein Equidae Anchitherium aurelianense Bollinger (1992) 50 51 Crêt-du-Locle 52 Age: ⁓13.6 Ma (van der Made, 2012) 53 Taxa: Locality Supraspecific taxa Species References Crêt du Locle Bovidae Tethytragus stehlini Thenius (1951) 54 55 Anwil 56 Age:13.37–13.17 Ma (Böhme et al., 2008) 57 Taxa: Locality Supraspecific taxa Species References Anwil Equidae Anchitherium aurelianense Engesser (1972) Anwil Palaeomerycidae Palaeomeryx sp. Engesser (1972) Pickford and Laurent Anwil Suidae Retroporcus matritensis (2014) 58 59 HUNGARY 60 Rudabánya 61 Age: 10.3–9.9 Ma (Böhme et al., 2008) 62 Taxa: Locality Supraspecific taxa Species References Hoploaceratherium Rudabánya Rhinocerotidae Heissig (2005) belvederense 63 64 POLAND 65 Opole 1 66 Age: 14.5–13.7 Ma (Harzhauser and Neubauer, 2018) 67 Taxa: Site Supraspecific taxa Species References 4 Opole 1 Proboscidea Gomphotherium angustidens Kowalski (1990) Opole 1 Hoploaceratherium Rhinocerotidae Kowalski (1990) tetradactylum Opole 1 Rhinocerotidae Brachypotherium brachypus Kowalski (1990) Opole 1 Chalicotheriidae Anisodon grande Kowalski (1990) Opole 1 Equidae Anchitherium aurelianense Kowalski (1990) Opole 1 Palaeomerycidae Palaeomeryx eminens Kowalski (1990) 68 69 Przeworno (1+2) 70 Age: ⁓12.2 Ma (Aiglstorfer et al., 2014b) 71 Taxa: Site Supraspecific taxa Species References Przeworno (1+2) Proboscidea Gomphotherium angustidens Kowalski (1990) Przeworno (1+2) Rhinocerotidae Brachypotherium brachypus Kowalski (1990) Przeworno (1+2) Equidae Anchitherium aurelianense Kowalski (1990) 72 73 GERMANY 74 Eggingen-Mittelhart 3 75 Age: 17.9–17.85 Ma (Böhme et al., 2012b) 76 Taxa: Site Supraspecific taxa Species References Eggingen – Sach and Heizmann Proboscidea Prodeinotherium bavaricum Mittelhart 3 (2001) Eggingen – Sach and Heizmann Proboscidea Gomphotherium angustidens Mittelhart 3 (2001) Eggingen – Sach and Heizmann Palaeomerycidae Palaeomeryx kaupi Mittelhart 3 (2001) Eggingen – Sach and Heizmann Carnivora Amphicyon cf. major Mittelhart 3 (2001) 77 78 Langenau 79 Age: 17.8–17.75 Ma (Böhme et al., 2012b) 80 Taxa from: Site Supraspecific taxa Species References Sach and Heizmann Langenau 1 Proboscidea Prodeinotherium bavaricum (2001) Langenau 1 Proboscidea Gomphotherium angustidens Sach and Heizmann 5 (2001) cf. Brachypotherium Sach and Heizmann Langenau 1 Rhinocerotidae brachypus (2001) Sach and Heizmann Langenau 1 Equidae Anchitherium aurelianense (2001) Sach and Heizmann Langenau 2 Palaeomerycidae Palaeomeryx sp. (2001) Langenau 1 Cervidae Heteroprox larteti Böhme et al. (2012a) 81 82 Gerlenhofen 83 Age: ⁓17 Ma (Aiglstorfer et al., 2014b) 84 Taxa: Locality Supraspecific taxa Species References Gomphotherium cf. Sach and Heizmann Gerlenhofen Proboscidea angustidens (2001) Sach and Heizmann Gerlenhofen Rhinocerotidae Brachypotherium brachypus (2001) 85 86 Sandelzhausen 87 Age: 16.58–16.37 Ma (Böhme et al., 2011) 88 Taxa: Locality Supraspecific taxa Species References Sandelzhausen Proboscidea Zygolophodon turicensis Göhlich (2010) Sandelzhausen Equidae Anchitherium aurelianense Kaiser (2009) Sandelzhausen Palaeomerycidae Germanomeryx fahlbushi Rössner (2010) 89 90 Häder 91 Age: ⁓16,2 Ma (Böhme et al., 2012a) 92 Taxa: Locality Supraspecific taxa Species References Häder Chalicotheriidae Anisodon grande Guérin 2012 Eronen and Rössner Häder Equidae Anchitherium aurelianense (2007) Eronen and Rössner Häder Cervidae Heteroprox larteti (2007) 93 94 Heggbach 6 95 Age: ⁓16 Ma (Aiglstorfer et al., 2014b) 96 Taxa: Locality Supraspecific taxa Species References Heggbach Proboscidea Gomphotherium angustidens Bechly and Sach (2002) Heggbach Equidae Anchitherium aurelianense Bechly and Sach (2002) Heggbach Palaeomerycidae Palaeomeryx kaupi Bechly and Sach (2002) Heggbach Cervidae Heteroprox larteti Böhme et al. (2012a) Heggbach Carnivora Amphicyon major Bechly and Sach (2002) 97 98 Stätzling 99 Age: ⁓14.8 Ma (Aiglstorfer et al., 2014b) 100 Taxa: Locality Supraspecific taxa Species References Stätzling Eronen and Rössner Rhinocerotidae Brachypotherium brachypus (2007) Stätzling Eronen and Rössner Chalicotheriidae Anisodon grande (2007) Stätzling Eronen and Rössner Equidae Anchitherium aurelianense (2007) Stätzling Eronen and Rössner Palaeomerycidae Palaeomeryx eminens (2007) Stätzling Cervidae Heteroprox larteti Böhme et al. (2012a) 101 102 Friedberg 103 Age: ⁓14.2 (Böhme et al., 2012a) 104 Taxa: Locality Supraspecific taxa Species References Friedberg Eronen and Rössner Rhinocerotidae Brachypotherium brachypus (2007) 105 106 Steinheim 107 Age: ⁓13.5 Ma (Tütken et al., 2006) 108 Taxa: Locality Supraspecific taxa Species References Steinheim Rhinocerotidae Brachypotherium brachypus Tütken et al. (2006) Steinheim Equidae Anchitherium aurelianense Tütken et al. (2006) Steinheim Palaeomerycidae Palaeomeryx eminens Tütken et al. (2006) 7 Steinheim Cervidae Heteroprox larteti Böhme et al. (2012a) 109 110 Massenhausen 111 Age: ⁓12.6 Ma (Böhme, 2012a) 112 Taxa: Locality Supraspecific taxa Species References Hoploaceratherium Eronen and Rössner Massenhausen Rhinocerotidae tetradactylum (2007) Massenhausen Eronen and Rössner Rhinocerotidae Brachypotherium brachypus (2007) Massenhausen Eronen and Rössner Chalicotheriidae Anisodon grande (2007) Massenhausen Eronen and Rössner Equidae Anchitherium aurelianense (2007) 113 114 Hammerschmiede 115 Age: 11.62–11.0 Ma (Böhme et al., 2008, 2019) 116 117 Taxa: Locality Supraspecific taxa Species References Hammerschmied Hoploaceratherium Rhinocerotidae Kirscher et al. (2016) e belvederense Hammerschmied Chalicotheriidae Anisodon grande Kirscher et al. (2016) e 118 119 Atzelsdorf 120 Age: ⁓11.1 Ma (Aiglstorfer et al., 2014b) 121 122 Taxa: Locality Supraspecific taxa Species References Atzelsdorf Equidae Anchitherium aurelianense Hillenbrand et al. (2009) Atzelsdorf Palaeomerycidae Palaeomeryx eminens Hillenbrand et al. (2009) 123 124 125 References 8 126 Aiglstorfer, M., Bocherens, H., Böhme M., 2014a. Large Mammal Ecology in the late Middle 127 Miocene lokality Gratkorn (Austria). Paleodivers. Paleoenv. 94, 189–213. 128 https://doi.org/10.1007/s12549-013-0145-5 129 Aiglstorfer, M., Rössner, G.E., Böhme, M., 2014b. Dorcatherium naui and pecoran ruminants 130 from the late Middle Miocene Gratkorn locality (Austria). The Sarmatian vertebrate 131 lokality Gratkorn, StyrianBasin. Paleodivers. Paleoenv. 94, 83–123. 132 https://doi.org/10.1007/s12549-013-0141-9 133 Astibia, H. 2012. Les Palaeomerycidae (Artiodactyla) de Sansan. In: Peigné, S., Sen, S. 134 (Eds.), Mammifères de Sansan. Mem. Mus. Natl. Hist. Nat. 203, 221–224. 135 Bechly, G., Sach, V.J., 2002. An interesting new fossil dragonfly (Anisoptera: Libellulidae: 136 “Brachydiplacini”) from the Miocene of Germany, with a discussion on the phylogeny of 137 Tetrathemistinae and a fossil list for the lokality. Heggbach 325, 1–11. 138 Böhme, M, Ilg, A., Winklhofer, M., 2008. Late Miocene “washhouse” climate in Europe. 139 Earth Planet. Sci. Lett. 275, 393–401. https://doi.org/10.1016/j.epsl.2008.09.011 140 Böhme, M., Winklhofer, M., Ilg A., 2011. Miocene precipitation in Europe: Temporal trends 141 and spatial gradients. Palaeogeogr. Palaeoclim. Palaeoecol. 304, 212–218. 142 https://doi.org/10.1016/j.palaeo.2010.09.028 143 Böhme, M., Aiglstorfer, M., Uhl, D., Kullmer, O., 2012a. The antiquity of the Rhine River: 144 stratigraphic coverage of the Dinotheriensande (Eppelsheim Formation) of the Mainz 145 Basin (Germany). PLoS One 7, e36817. https://doi.org/10.1371/journal.pone.0036817 146 Böhme, M., Vasilyan, D., Winklhofer, M., 2012b. Habitat tracking, range dynamics and 147 palaeoclimatic significance of Eurasian giant salamanders (Cryptobranchidae) –indications 148 for elevated Central Asian humidity during Cenozoic global warm periods. Palaeogeogr. 149 Palaeoclim. Palaeoecol. 342/343, 64–72. https://doi.org/10.1016/j.palaeo.2012.04.032 9 150 Böhme, M., Spassov, N., Fuss, J., Tröscher, A., Deane, A. S., Prieto, J., Kirscher, U., 151 Lechner, T., Begun, D.R., 2019. A new Miocene ape and locomotion in the ancestor of 152 great apes and humans. Nature 575, 489–493. https://doi.org/10.1038/s41586-019-1731-0 153 Bollinger, T.H., 1992. Kleinsäugerstratigraphie der miozänen Hörnlischüttung (Ostschweiz). 154 Documentanaturae 75, 1–296. 155 Daxner-Höck, P.M., Miklas-Tempfer, U.B., Göhlich, K., Hüttunen, E., Kazar, D., Nagel, 156 G.E., Roessner, O., Schultz, Ziegler, R., 2004. Marine and terrestrial vertebrates from the 157 Middle Miocene of Grund (Lower Austria). Geol. Carpath. 55, 191–197. 158 Engesser, B., 1972. Die obermiozäne Säugetierfauna von Anwil (Baselland). Tätigk. 159 Naturforsche. Ges. Baselland 28, 35–363. 160 Eronen, J.T., Rössner, G.E., 2007. Wetland Paradise Lost: Miocene Community Dynamics in 161 Large Herbivore Mammals from the German Molasse Basin. Evol. Ecol. Res. 9, 471– 162 494. 163 Fejfar, O., 1990. The Neogene vertebrate paleontology sites of Czechoslovakia: A 164 contribution to the Neogene terrestric biostratigraphy of Europe based on rodents, in: 165 Lindsay, E.H., Fahlbusch, V., Mein P. (Eds.), Proceedings of a NATO Advances Research 166 Workshop on European Neogene Mammal Chronology. Plenum Press, New York, pp. 167 211–236. 168 Heissig, K., 2012. Les Rhinocerotidae (Perissodactyla) de Sansan. In: Peigné, S., Sen, S. 169 (Eds.), Mammifères de Sansan. Mem. Mus. Natl. Hist. Nat. 203, 317–485. 170 Hillenbrand, V., Göhlich U.B., Rössner, G.E., 2009. The early Vallesian vertebrates of 171 Atzelsdorf (Late Miocene, Austria). Ruminantia. Ann. Naturhist. Mus. Wien 111 A, 519– 172 556. 10 173 Gross, M., Böhme, M., Havlik, P., Aiglstorfer, M., 2014. The late Middle Miocene 174 (Sarmatian s.str.) fossil site Gratkorn – the first decade of research, geology, stratigraphy 175 and vertebrate fauna. Paleodivers. Paleoenv. 94, 5–20. 176 https://doi.org/10.1007/s12549-013-0149-1 177 Göhlich U.B., 2010. The Proboscidea (Mammalia) from the Miocene of Sandelzhausen 178 (southern Germany). Paläontolo. Z. 84, 163–204. 179 https://doi.org/10.1007/s12542-010-0053-1 180 Guérin C., 2012. Anisodon grande (Perissodactyla, Chalicotheriidae) de Sansan. In: Peigné, 181 S., Sen, S. (Eds.), Mammifères de Sansan. Mem. Mus. Natl. Hist. Nat. 203, 279–315. 182 Harzhauser, M., Neubauer, T.A., 2018. Opole (Poland) – a key locality for middle Miocene 183 terrestrial mollusc faunas. Bull. Geosci. 93, 71–146. 184 https://doi.org/10.3140/bull.geosci.1692 185 Heissig, K., 2005. The fossil rhinoceroses of Rudabánya. Palaeontogr. Ital. 90, 217–2158. 186 Hyžný, M., Hudáčková, N., Biskupič, R., Rybár, S., Fuksi, T., Halásová, E., Zágoršek, K., 187 Jamrich, J., Ledvák, P., 2012. Devínska Kobyla – a window into the Middle Miocene 188 shallow-water marine environments of the Central Paratethys (Vienna Basin, Slovakia). 189 Acta Geol. Slovaca 4, 95–111. 190 Ivanov, M., in press. Miocene snakes of Eurasia - a review of the evolution of snake 191 communities, in: D. Gower and H. Zaher (eds.), A Contribution to the Origin and Early 192 Evolution of Snakes, Systematics Association Special Volume Series, Cambridge 193 University Press. 194 Kaiser, T.M., 2009. Anchitherium aurelianense (Equidae, Mammalia): a brachydont ‘‘dirty 195 browser’’ in the community of herbivorous large mammals from Sandelzhausen (Miocene, 196 Germany). Paläontol. Z. 83, 131–140. 197 https://doi.org/10.1007/s12542-009-0002-z 11 198 Kirscher, T., Prieto, J., Bachtadse, V., Abdul Aziz, H., Doppler, G., Hagmaier, M., Böhme 199 M., 2016. A biochronologic tie-point for the base of the Tortonian stage in European 200 terrestrial settings: Magnetostratigraphy of the topmost Upper Freshwater Molasse 201 sediments of the North Alpine Foreland Basin in Bavaria (Germany). Newsl. Stratigr. 49, 202 445–467. https://doi.org/10.1127/nos/2016/0288 203 Kowalski, K., 1990. Stratigraphy of Neogene mammals in Poland, in: Lindsay, E.H., 204 Fahlbusch, V., Mein, P. (Eds.), European Neogene Mammal Chronology. Plenum Press, 205 New York, pp. 193–209. 206 Pickford, M., Laurent, Y., 2014. Valorisation of palaeontological collections: nomination of a 207 lectotype for Conohyus simorrensis (Lartet, 1851), Villefranche d’Astarac, France, and 208 description of a new genus of tetraconodont. Estud. Geol. 70, 1–28. 209 http://dx.doi.org/10.3989/egeol.41261.262 210 Rössner, G., 2010. Systematics and palaeoecology of Ruminantia (Artiodactyla, Mammalia) 211 from the Miocene of Sandelzhausen (southern Germany, Northern Alpine Foreland Basin). 212 Paläontol. Z. 84, 123–162. https://doi.org/10.1007/s12542-010-0052-2 213 Sabol, M., Holec, P., 2002. Temporal and spatial distribution of Miocene mammals in the 214 Western Carpathians. Geol. Carpath. 53, 269–279. 215 Sabol, M., Kovac, M., 2006. Badenian Palaeoenvironment, Faunal Succession and 216 Biostratigraphy: A Case Study from Northern Vienna Basin, Devinska Nova Ves-Bonanza 217 site (Western Carpathians, Slovakia). Beitr. Palaont. 30, 415–425. 218 Sach, V.J., Heizmann, E.P.J., 2001. Stratigraphy and mammal faunas of the Brackwasser 219 molasse in the surroundings of Ulm (Southwest Germany). Stuttg. Beitr. Naturk Serie B, 220 310, 1–95. 12 221 Thenius, E. 1948. Zur Kenntnis der fossilen Hirsche des Wiener Beckens, unter besonderer 222 Berücksichtigung ihrer stratigraphischen Bedeutung. Ann. Nat. Hist. Mus. Wien. 56, 262– 223 308. 224 Thenius, E. 1951. Gazella cf. deperdita aus dem mitteleuropäischen Vindobonien und das 225 Auftreten der Hipparionfauna. Eclogae Geol. Helv. 44, 381–394. 226 Thenius, E., 1959. Wirbeltierfaunen, in: Lotze, F. (Ed.), Handbuch der Stratigraphischen 227 Geologie, III Band 2. Teil, Papp, A., Thenius, E., Tertiär. Ferdinand Enke Verlag, 228 Stuttgart, 328 pp. 229 Tütken, T., Vennemann. T.W., Janz, H., Heizmann, E.P.J., 2006. Palaeoenvironment and 230 palaeoclimate of the Middle Miocene lake in the Steinheim basin, SW Germany: A 231 reconstruction from C, O, and Sr isotopes of fossil remains. Palaeogeogr. Palaeoclimatol. 232 Palaeoecol. 241, 457–491. https://doi.org/10.1016/j.palaeo.2006.04.007 233 Van der Made, J., 2012. Eotragus clavatus (Artiodactyla, Bovidae, Boselaphini) de Sansan, 234 in: Peigné, S., Sen S. (Eds), Mammifères de Sansan. Mém. Mus. Nat. Hist. Natl. 203, 145– 235 199. 13 Supplementary Table 1 List of studied material with distribution of skeletal elements for each determined taxa. MINIMAL NUMBER OF ELEMENTS (MNE) SKULL UPPER DENTITION LOWER DENTITION Supraspecific taxa Species Institution Specimens Rounded bones Fresh bones Skull bones Antler/Horn Upper jaw Lower jaw I2 C P3 P4 M1 M2 M3 d4 p2 p3 Proboscidea Prodeinotherium bavaricum MZM 1 0 1 1 Proboscidea Zygolophodon turicensis MZM, IPUW2 ,RMM3 70 2 68 2 4, 1 2 27, 1 3 1 5 7, 1 5, 2 2 1 2 Proboscidea Gomphotherium angustidens MZM 4 0 4 1 1 2 Rhinocerotidae Hoploaceratherium sp. MZM 17 0 17 Rhinocerotidae Brachypotherium brachypus MZM, IPUW2 ,RMM3 24 1 23 1 2, 1 2 , 1 3 1 1 1 2 1 1 1 1 Chalicotheriidae Anisodon grande MZM 3 0 3 Equidae Anchitherium sp. MZM 3 1 2 1 Palaeomerycidae Palaeomerycidae indet. MZM, IPUW2 14 2 12 2 Cervidae Heteroprox larteti MZM 3 2 1 3 Bovidae Tethytragus stehlini MZM, IPUW2 20 7 13 1 11, 2 2 2 4 Suidae Retroporcus matritensis RMM3 1 0 1 13 Carnivora Amphicyon cf. major MZM, ÚGV1 3 1 2 1, 1 1 Testudinidae Testudo kalksburgensis MZM 8 0 8 Trionychidae Trionychidae indet. IPUW2 1 1 0 Cervidae/Bovidae Undetermined MZM, IPUW2 11 4 7 Proboscidea Undetermined MZM, IPUW2 190 20 170 2 Rhinocerotidae Undetermined MZM, IPUW2 34 1 33 1 3 Testudines Undetermined fragments MZM, IPUW2 57 14 43 TOTAL (MNE) 464 56 408 4 16 6 20 30 2 1 2 7 11 10 1 1 1 1 = Department of Geological Sciences, Faculty of Sciences, Masaryk University (ÚGV) 2 = Paleontological Institut University Vienna (IPUW) 3 = Regional Museum in Mikulov (RMM)