An Updated Review of The Quaternary Hippopotamus Fossil Records from the Iberian Peninsula

: This work presents a comprehensive review of the Quaternary fossil records of hippopota-muses from the Iberian Peninsula, unveiling biogeographical insights of global significance. The results presented herein include the inference of a delayed arrival of Hippopotamus ( Hippopotamus antiquus ) populations onto the Iberian Peninsula compared to other European Mediterranean regions, with an estimated age of ca. 1.7 Ma, in contrast to 2.1–2.2 Ma elsewhere. Moreover, we hypothesize the possibility of a short-lived coexistence between H. antiquus and Hippopotamus amphibius , close to the extinction of the former taxon (ca. 0.45 Ma). The local extinction of all hippopotamus populations on the Iberian Peninsula between MIS 5 and 3 is suggested here, mirroring proposals made for the Italian peninsula. Notable aspects of this fossil record include the abundance of specimens, previously undocumented anatomical elements, and partially complete individuals with articulated body segments. The remains analyzed herein also present different ontogenetic stages and sexual dimorphism. Moreover, the presence of specimens displaying paleopathologies provides valuable insights into ethological and paleoecological studies. The exceptional record of at least three events of human exploitation of hippopotamus stands out, with this being a rarity in the broader context of the archaeological and paleontological record of the European continent.


Introduction
Despite their putative restrictive environmental necessities, fossil records suggest that Hippopotamus inhabited continental Europe for at least 2 million years during the Quaternary period [1][2][3].According to the current data, changes in the African and European ecosystems as well as in their interconnecting areas allowed the passage and the following establishment of hippopotamus populations between these zones, forming a complex phylogeographical context [4].In order to unravel these paleobiological networks, approaches at different scales are necessary, from a global approach that interrelates existing regional data to the evaluation of specific cases that provide a geographical and chronological basis for biological information.In the case of hippopotamuses, there has been an inertia in this type of work in the last few decades.Since the end of the last century, different studies focused on hippopotamus occurrences have been carried out.These works include Stuart and Gibbard [5] for the United Kingdom, Pandolfi and Petronio [6] and Martino and Pandolfi [2] for the Italian peninsula, Athanassiou [7] for Greece, and Martino et al. [8] for the Portuguese record.
During the Quaternary, at least two hippopotamus dispersals into continental Europe are inferred, both from taxa belonging to the genus Hippopotamus [9].Hippopotamus antiquus is a species erected on specimens from the Upper Valdarno (Tuscany, Italy) ( [2] and references therein).The first appearance datum (FAD) of H. antiquus can probably be traced back to the Middle-Late Villafranchian boundary, probably before 2 Ma [2,10].Its origin is still disputed, although it seems closely related to the large African species Hippopotamus gorgops [11].H. antiquus has a wide paleogeographic distribution range, extending from the Iberian Peninsula to Georgia and reaching the United Kingdom and the entire Italian Peninsula [2,5,8,12].The last records of this taxon in Europe are referred to in the mid-Middle Pleistocene and are rather conflicting [8], with the taxon Hippopotamus tiberinus (=H.ex gr.H. antiquus in [13]) being considered to encompass the last forms of this lineage or it could represent a new interaction with African populations (see [4,14]).
More limited are the fossil records of Hippopotamus amphibius, and its dispersal towards the European continent is still poorly documented [8].However, it is generally considered that between 0.5-0.4Ma, specimens belonging to this species were present in Europe [15].This chronological range would allow a presumed co-occurrence of H. amphibius with H. antiquus in areas such as the Italian peninsula [2].Even so, the specific discrimination of most of the skeletal elements of both species is complex.Some complete skulls and postcranial material of H. amphibius were collected from the English locality of Barrington [16], and a skull assigned to H. amphibius was described from the Italian locality of Tor di Quinto [15,17].The last recorded occurrences of this taxon go as far as MIS 5-MIS 3 in Italy [6], without a clear assessment of the conditions that led to its extinction across Europe.Probably in Greece, H. amphibius survived until ca 30 ka since its presence is reported from Dyrós Cave [7].An analysis of the fossil and archaeological records of the Levantine corridor shows that the last populations of hippopotamuses persisted in this area until historical times [18].
The long paleontological and archaeological research traditions of the Iberian Peninsula have favored a rigorous method for excavation and contextualization of the fossil and archaeological records.This, together with the presence of large basins with Quaternary sedimentary deposits and areas with karstic activity [19], creates very favorable conditions to house a fossil record of great interest.A better understanding of the spatiotemporal distribution of hippopotamuses on the Iberian Peninsula is necessary to improve the paleoenvironmental interpretation of their habitats at times of coexistence with human populations and to better understand how their ecosystems changed.Under these premises, this paper compiles, re-evaluates, and provides new data on the fossil record of hippopotamuses on the Iberian Peninsula during the Quaternary.Special attention is paid to paleobiological parameters, also contextually including the relevant archaeological information.

Materials and Methods
In order to carry out an exhaustive update of the Iberian fossil records of hippopotamuses, a four-stage methodology was applied: 1.
Compilation and consultation of all of the available bibliographies on hippopotamus fossils on the Iberian Peninsula and the sites where they have been found.The key references for each case are listed in Table S1.

2.
Direct consultation with the researchers responsible for excavations on the findings.Data from this type of consultation are referenced as "pers.comm.".The large volume of fossil material evaluated (either directly, bibliographically or by personal communication) cannot be described in detail in this paper.Nevertheless, in the following sections, some of the specimens and fossil assemblages with the greatest paleobiological, archaeological, or heritage value will be highlighted.In turn, the fossil assemblages are summarized in the data associated with the locality or archaeo-paleontological level of the find (Table S1).These sites and geological layers will be treated as cases for the analysis of the fossil records.
The taxonomic determinations of the hippopotamus specimens included were reviewed, accepting those determinations that are sufficiently justified and based on discriminating anatomical elements.In this work, we considered the taxonomy proposed by Petronio [9]; we therefore accepted the validity of H. antiquus (=H.major) and H. amphibius, and we avoided using H. tiberinus, H. ex gr.antiquus, or H. incognitus.
The most current geochronology and stratigraphic framework proposed for each find included has been respected.Geochronologic ages are given in thousand or million years before the present, abbreviated as "ka" or "Ma", respectively.

Results
The review carried out in this work highlights the appraisal of at least forty-eight different localities or deposits with the occurrence of hippopotamuses on the Iberian Peninsula.Most of these localities are distributed parallel to the Mediterranean coast, although there are records from the center of the peninsula, the western part of the Atlantic coast, and a few specimens from the northern part of the peninsula (Figure 1).
A total of twenty-five deposits with hippopotamuses have been dated to the Early Pleistocene, while fifteen belong to the Middle Pleistocene and only two belong to the Late Pleistocene (Figure 2).Six of the cases in our dataset could not be assigned to a more precise chronology than the undifferentiated Pleistocene.The earliest record found on the Iberian Peninsula, the Mencal-9 site, is associated with a chronology of ca.1.7 Ma (Figure 3j).On the other hand, the last records of hippopotamuses on the peninsula are ascribed to the Late Pleistocene, between MIS 5 and MIS 3, at the sites of Cueva del Toll and Sima de las Palomas de Cabezo Gordo (Figure 3k).If we focus on the taxa considered, the oldest specimens reliably determined as H. antiquus are found at Venta Micena (1.6-1.4Ma), and the latest are found at Condeixa (ca.0.45 Ma).According to the records from Solana de Zamborino and Las Jarillas, H. amphibius may have made its appearance in the records 0.48-0.4Ma ago, surviving until the Late Pleistocene (Figure 2).
If we evaluate the geographic and chronological data of all hippopotamus occurrences, we can observe a wide distribution area of hippopotamuses in the Early Pleistocene that would extend from the southwest of the peninsula (Algoz) and the eastern Pyrenees (Incarcal Complex and Bòbila Ordis) to the center-north of the peninsula (Atapuerca Complex), including the entire Mediterranean coast (Figure 1).During the Middle Pleistocene, this extension remained constant, increasing the records in the center of the peninsula and in the central part of the Atlantic coast.At the end of the Middle Pleistocene and during the Late Pleistocene, hippopotamus records were much more limited and only located on the Mediterranean coast (Figure 1).A total of twenty-five deposits with hippopotamuses have been dated to the Early Pleistocene, while fifteen belong to the Middle Pleistocene and only two belong to the Late Pleistocene (Figure 2).Six of the cases in our dataset could not be assigned to a more precise chronology than the undifferentiated Pleistocene.The earliest record found on the Iberian Peninsula, the Mencal-9 site, is associated with a chronology of ca.1.7 Ma (Figure 3j).On the other hand, the last records of hippopotamuses on the peninsula are ascribed to the Late Pleistocene, between MIS 5 and MIS 3, at the sites of Cueva del Toll and Sima de las Palomas de Cabezo Gordo (Figure 3k).If we focus on the taxa considered, the oldest specimens reliably determined as H. antiquus are found at Venta Micena (1.6-1.4Ma), and the latest are found at Condeixa (ca.0.45 Ma).According to the records from Solana de Zamborino and Las Jarillas, H. amphibius may have made its appearance in the records 0.48-0.4Ma ago, surviving until the Late Pleistocene (Figure 2).Among the most important features of these fossil records, there is a great number of hippopotamus specimens from sites such as Vallparadís Estació, with more than 3000 fossils.The studied fossil assemblages contain several complete or partial skulls and mandibles (e.g., the Cal Guardiola Lower Unit, La Mina, Bòbila Ordis, the Vallparadís Estació Upper Unit, Incarcal-I, or Condeixa; Figure 3a-i) and partial skeletons (e.g., the Cal Guardiola Lower Unit or Incarcal-I; Figure 3n) including anatomical element such as the hyoid bone (e.g., Incarcal-I), together with specimens belonging to different age classes and sexes (e.g., the Vallparadís Section; Figure 3a-d).In addition to valuable data for the autoecological analysis of the hippopotamus populations, at some sites, we found information on the biotic (plant and faunal) and abiotic context in which these populations lived (e.g., Barranc de la Boella [10]).In some of the localities, direct interaction of other organisms with hippopotamuses was detected, such as marks on the bones of exploitation by carnivores (e.g., Barranco León [20] or the Vallparadís Section [21]; Figure 3m) or cutmarks by humans (Barranco León, Fuente Nueva-3 [20], and Cueva de Bolomor [22]; Figure 3l).If we evaluate the geographic and chronological data of all hippopotamus occurrences, we can observe a wide distribution area of hippopotamuses in the Early Pleistocene that would extend from the southwest of the peninsula (Algoz) and the eastern Pyrenees (Incarcal Complex and Bòbila Ordis) to the center-north of the peninsula (Atapuerca Complex), including the entire Mediterranean coast (Figure 1).During the Middle Pleistocene, this extension remained constant, increasing the records in the center of the peninsula and in the central part of the Atlantic coast.At the end of the Middle Pleistocene and during the Late Pleistocene, hippopotamus records were much more limited and only located on the Mediterranean coast (Figure 1).
Among the most important features of these fossil records, there is a great number of hippopotamus specimens from sites such as Vallparadís Estació, with more than 3000 fossils.The studied fossil assemblages contain several complete or partial skulls and mandibles (e.g., the Cal Guardiola Lower Unit, La Mina, Bòbila Ordis, the Vallparadís Estació Upper Unit, Incarcal-I, or Condeixa; Figure 3a-i) and partial skeletons (e.g., the Cal Guardiola Lower Unit or Incarcal-I; Figure 3n) including anatomical element such as the hyoid bone (e.g., Incarcal-I), together with specimens belonging to different age classes and sexes (e.g., the Vallparadís Section; Figure 3a-d).In addition to valuable data for the autoecological analysis of the hippopotamus populations, at some sites, we found information on the biotic (plant and faunal) and abiotic context in which these populations lived (e.g., Barranc de la Boella [10]).In some of the localities, direct interaction of other organisms with hippopotamuses was detected, such as marks on the bones of exploitation by carnivores (e.g., Barranco León [20] or the Vallparadís Section [21]; Figure 3m) or cutmarks by humans (Barranco León, Fuente Nueva-3 [20], and Cueva de Bolomor [22]; Figure 3l).(i) mandible of H. antiquus found in Condeixa, assigned to the last recorded presence of the species on the Iberian Peninsula [8]; (j) fragment of a molar from Mencal-9, assigned to the oldest recorded presence of Hippopotamus on the Iberian Peninsula (Arribas, pers.comm.);(k) incisor fragment determined as Hippopotamus amphibius found in the Sima de las Palomas de Cabezo Gordo, assigned to one of the last recorded presences of hippopotamuses on the Iberian Peninsula (Walter, pers.comm.);(l) cut mark reported on a bone of antiquus at the locality of Fuente Nueva (modified from [20]); (m) femur of H. antiquus with marks of exploitation by hyenas found at Fuente Nueva (modified from [20]); and (n) partial skeleton of H. antiquus found at Cal Guardiola.Scale bar (a-i) 3 cm, (j-k) 1 cm and (n) 10 cm.

Phylogenetic and Biogeographical Relationships of Hippopotamus of the Iberian Peninsula
Considering the data obtained, the oldest chronological record of specimens assignable to the genus Hippopotamus on the Iberian Peninsula was found at the Mencal-9 site.Despite the fragmentary nature of the remains found at Mencal-9 (two fragmented incisors (M9-2009-0012 and M9-2009-0059) and a fragmented molar [Figure 3j]), their taxonomic attribution to the genus Hippopotamus is reliable (Garrido and Arribas, pers.comm.).Chronologically, the Mencal-9 deposits have been dated to ca. 1.7 Ma by magnetostratigraphic and biochronological techniques [23] (Garrido and Arribas, pers.comm.).Some interesting specimens originally assigned to H. antiquus were also found in Venta Micena layers dated ca.1.6-1.4Ma [24].Possibly, the remains from these two localities represent some of the earliest populations of H. antiquus that would have entered the Iberian Peninsula slightly later than in other Mediterranean areas: ca. 2 Ma in Greece (Elis [25]), ca.2.2-2.1 Ma in Italy (e.g., Coste San Giacomo [26]), or ca.2.2-2.1 Ma in France (Senèze [27]; J.M.-M.unpublished data).According to the hypothesis of van der Made et al. [14], these populations could correspond to an initial influx of African forms related to the controversial taxon Hippopotamus kaisensis (referred to as H. antiquus in Europe).Subsequently, these populations would have disappeared, leading to a period without hippopotamuses in Europe, which would have probably ended ca.1.4 Ma ago with the arrival of populations of Hippopotamus gorgops (referred to as Hippopotamus tiberinus in Europe).However, the relatively close chronologies of the Barranco León site (ca.1.4-1.2Ma), where hippopotamus fossils have also been found [28], and the general scarcity of fossil sites from this time span (1.6-1.4Ma) on the Iberian Peninsula [29] do not particularly support the previously mentioned hypothesis.
The number of hippo sites increased enormously during the so-called 'Early-Middle Pleistocene Transition', from the end of the Jaramillo subchron to the end of the Early Pleistocene.In this chronological timespan, H. antiquus populations extended from the southwesternmost tip of the peninsula (Algoz [30]) to its northeastern boundary (Incarcal Complex and Bòbila Ordis [31,32]).This increase in the fossil records of hippopotamuses is also shared in other areas of Europe, such as the Italian peninsula [2], where fossil assemblages, such as Collecurti, with more than five hundred specimens of this taxon were described by Mazza and Ventra [33].With the transition to the Middle Pleistocene, the abundance of fossil sites with hippopotamus specimens decreased again.Indeed, Mazza [4] recognized a phenotypic trend in H. antiquus populations from ca. 1 Ma onwards that described as a new species Hippopotamus tiberinus.The validity of this taxon is disputed but a discussion on this topic is beyond the scope of this paper.However, it should be noted that van der Made et al. [34] noted a certain affinity between the specimens found in level TD8 from Gran Dolina (Atapuerca) and the specimens determined by Mazza [4] as H. tiberinus.
The last reliable occurrences of H. antiquus in Europe were recorded in the mid-Middle Pleistocene.Martino et al. [8] re-evaluated the specimens collected from Condeixa dated to ca. 0.45 Ma and confirmed their attribution to H. antiquus.If we consider as the earliest occurrence of H. amphibius in Europe the record from Tor di Quinto (most likely Cava Montanari [17]) dated ca.0.55-0.45Ma [15], a brief overlap of both taxa in Europe seems likely.On the Iberian Peninsula, there is no record of H. amphibius sufficiently informative to be sure of the chronological range of its presence.However, specimens from Solana de Zamborino (0.48-0.3 Ma [35]) and Las Jarillas (0.45-0.4 Ma [36]) have been determined as H. amphibius.If these determinations are confirmed, we should consider the possibility of the coexistence of H. antiquus and H. amphibius in a limited geographical area as the south and west of the Iberian Peninsula.
During the second half of the Middle Pleistocene, there are few sites where hippopotamuses are present, but unfortunately, their record is rather fragmentary.Despite this, the known finds point to the greatest geographical extension of this taxon, from the centralwest (Condeixa and Melhada [8]) to the northeast (Can Deu [37]) of the peninsula.The group of fluvial terrace sites on the central and southern peninsula also highlight a constant presence of H. cf.amphibius populations until the end of the Middle Pleistocene (Áridos-1 [38], Campo de Tiro [39], Buenavista Inferior [39], or Las Jarillas [36]).A record with a slightly higher abundance of this taxon was found at the end of the Middle Pleistocene (MIS 5e) in the Bolomor Cave site [40] (Fernández Peris and Blasco, pers.comm.), although a detailed paleobiological study of these materials has not yet been carried out.These time periods coincide with those proposed for the findings of the largest accumulation of H. amphibius specimens at the Barrington locality (UK, ca.0.13-0.115Ma [16]).Faure [41], based on the English specimens, erected the disputed taxon Hippopotamus incognitus, listed as invalid by Petronio [9].
The local extinction of H. amphibius is poorly recorded on the Iberian Peninsula.The most recent findings of this taxon point to its last occurrences around MIS 4.Even so, these findings present a very limited stratigraphic and chronological contextualization.The specimens cited by Donner and Kurtén [42] at the Toll Cave site, and dated to MIS 4, have not been located in subsequent revisions (Blasco, pers. comm.).On the other hand, the specimen cited at the Sima de las Palomas de Cabezo Gordo site was found thirty years ago in the Upper Cuttin, without any stratigraphic context (Walker, pers.comm.).Estimation of the original stratigraphic position of the specimen has allowed us to assess a chronological range for its deposition between MIS 5 and MIS 3 [43].These last local presences' chronologies also approximate the timing of the extinction of H. amphibius populations across Europe.Although specimens from the Italian site of Grotta Romanelli dated to MIS 3 were considered the last occurrence of hippopotamuses in continental Europe by Martino and Pandolfi [2], recently, Pieruccini et al. [44] re-dated the IUS3 to MI5e.

Paleoecological Considerations
Although the niche parameters of H. antiquus are still under discussion [10,45,46], the need for permanent bodies of water for its survival [47] and the presence of more or less open areas with grasslands [46] seem a clear requirement.Although these inferences are not very precise, they provide information on the minimum climatic parameters for its arrival in Europe and, specifically, for its dispersal across the Iberian Peninsula.The extension of the distribution area from Africa to the Iberian Peninsula in the middle of the Early Pleistocene raises two possible scenarios: (1) the presence of sufficient water bodies and an orographically traversable terrain for its passage through the Levantine corridor and the crossing of the entire European continent or (2) the reconsideration of a possible aquatic entry from North Africa through the western Mediterranean.
Once the populations of H. antiquus were established on the Iberian Peninsula, a continuous period of more than one million years of presence in this area was observed.The stability of these populations on a large part of the peninsula is surprising, as climatic conditions were progressively becoming harsher.These changes, known as the 'Early-Middle Pleistocene Transition', were characterized by a gradual increase in the severity and duration of glacial cycles, along with an increase in seasonality and aridity [48,49].Equally surprising is the possibility of a very brief coexistence of H. antiquus and H. amphibius in a restricted geographical area in the mid-Middle Pleistocene.This phenomenon has been previously proposed for the Italian peninsula as well [2].More information on the niche differences between the two hippopotamuses is needed to assess the possible role of H. amphibius in the extinction of H. antiquus.The same can be said regarding the environmental conditions that may have led to a shift from the European to the extant hippopotamus, as it has been hypothesized that both taxa had the same dietary and climatic constraints.
The Middle Pleistocene is characterized by the presence of rapid cyclicity between glacial and interglacial intervals [50].The arrival into the Northern Hemisphere of H. amphibius, classically associated with warm and wet conditions, is remarkable [47].Even more surprising is its persistence in Europe until the Late Pleistocene, with an apparent continuous record on both the Italian peninsula [2] and the Iberian Peninsula.As has already been proposed for other taxa (e.g., [51,52]; among others), the population mobility dynamics of hippopotamuses during the Middle Pleistocene could correspond to a pushpull mechanism, lingering in glacial intervals in these warmer peninsulas and dispersing to more northern areas during interglacial intervals (e.g., UK [5,16,47]).However, a better understanding of the Central European and British records is crucial for the evaluation of these biogeographical hypotheses.Similarly, a detailed study of the state of health of the last well-recorded populations of H. amphibius in Europe (e.g., Barrington [16] or Bolomor [22]) would allow us to refine the causes of the continent-wide disappearance of a group of mammals that inhabited this area for more than two million years.
In particular, the Iberian Quaternary hippopotamus records offer exceptional specimens and fossil assemblages of great importance in paleobiological studies.Localities such as Incarcal-I (ca.0.86 Ma; MIS21 [31,53]) have anatomical elements with an extremely limited representation in fossil records.In the same locality, together with the Vallparadís Section fossil assemblage (1.2-0.86Ma; [21,54,55]), partially complete individuals have been excavated, with most of their anatomical elements articulated (Figure 3n).At the same time, at these sites, we found individuals characterized by different ontogenetic stages and belonging to both sexes (Figure 3a-d).Another highlight of this fossil record is the presence of different skeletal pathologies (e.g., dental enamel hypoplasias, malocclusion, stress deformities, or joint problems) that are being studied in a global bioclimatic context [49].
Most of the fossil assemblages with hippopotamus presence on the Iberian Peninsula are the result of interspecific interactions (e.g., accumulation by carnivores [56]) or show evidence of this type of event (e.g., feeding by hyenas [21,57]; Figure 3m).The peculiar hippopotamus lifestyle brings a particular value to the characterization of interspecific relationships, allowing us to infer the behavior of carnivores around water bodies.In contrast, there is very little evidence of human exploitation of hippopotamuses across the European continent [58] (Figure 3l).Of this scarce evidence, only one reference is outside the Iberian Peninsula, in the Greek locality of Marathousa [58].In particular, a limited number of cut marks identified on specimens from the sites of Barranco León [59] and Fuente Nueva-3 [20] allow us to propose human access to hippopotamus carcasses in the Early Pleistocene.More evident seems to be the accumulation of hippopotamus remains in the locality of Bolomor Cave at the end of the Middle Pleistocene [22].Further elaboration on this topic is beyond the scope of this paper and will be dealt with in more detail in future works.

Conclusions
The review and detailed study of the Quaternary fossil record of hippopotamuses from the Iberian Peninsula has allowed us to highlight several biogeographical data of global importance: (a) we infer the arrival of Hippopotamus antiquus populations to the Iberian Peninsula later than to the rest of the European Mediterranean areas (ca.1.7 Ma vs. 2.1-2.2Ma); (b) the possibility of a brief coexistence between H. antiquus and Hippopotamus amphibius on the peninsula in chronologies close to the extinction of the populations of the former taxon (ca.0.45 Ma); and (c) the idea of a local extinction of all hippopotamus populations in this area between MIS 5 and MIS 3 is considered, mostly coinciding with the proposals published for the Italic peninsula.The biogeographical proposals considered have been evaluated in a paleoecological and paleoclimatic context, opening the possibility of new paleoenvironmental reconstructions.
Together with these data, the potential of the Iberian record for the progress of paleozoological and paleoecological studies of the European Quaternary, including aspects related to human evolution, has been highlighted: (a) the abundance of records, the presence of poorly-known anatomical elements and the occurrence of partially complete individuals displaying articulated body segments; (b) a good representation of intraspecific diversity, with individuals of different ontogenetic stages and assignable to both sexes; (c) the existence of specimens with paleopathologies particularly relevant for paleoecological reconstructions; (d) the exceptional presence of at least three events of hippopotamus exploitation by humans, scarcely documented on the whole European continent.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/quat7010004/s1,Table S1: Quaternary hippopotamus fossil record from the Iberian Peninsula.References  are cited in the Supplementary Materials

Figure 1 .
Figure 1.Schematic map of the Iberian Peninsula with the localities where Hippopotamus specimens have been found.In the lower part of the map, the minimum geographic distribution range that hippopotamus populations would reach in each sub-epoch according to their fossil records is highlighted.

Figure 1 .
Figure 1.Schematic map of the Iberian Peninsula with the localities where Hippopotamus specimens have been found.In the lower part of the map, the minimum geographic distribution range that hippopotamus populations would reach in each sub-epoch according to their fossil records is highlighted.

Figure 2 .
Figure 2. Chronological representation of Hippopotamus finds from the Iberian Peninsula and their taxonomic approximation.On the right, the estimated biochronological range of each species on the Iberian Peninsula is indicated with respect to its estimate for Europe as a whole.

Figure 2 .
Figure 2. Chronological representation of Hippopotamus finds from the Iberian Peninsula and their taxonomic approximation.On the right, the estimated biochronological range of each species on the Iberian Peninsula is indicated with respect to its estimate for Europe as a whole.

Figure 3 .
Figure 3.Some of the most outstanding finds of hippopotamuses on the Iberian Peninsula: (a-d) ontogenetic series of jaws of Hippopotamus antiquus from the Vallparadís Section, (a) infant jaw from Vallparadís Estació (EVT; IPS99003), (b) juvenile jaw from EVT (IPS127115), (c) young adult jaw from EVT (IPS127243), and (d) adult jaw from Cal Guardiola (IPS14513; (e) skull of H. antiquus from Cal Guardiola (IPS14960); (f) skull of H. antiquus compressed dorso-ventrally found in the Vallparadís Estació layer EVT7 (IPS127242); (g) reconstruction of the skull of H. antiquus found in Bòbila Ordis; (h) reconstruction of the skull of H. antiquus found in Barranc de la Boella[10]; (i) mandible of H. antiquus found in Condeixa, assigned to the last recorded presence of the species on the Iberian Peninsula[8]; (j) fragment of a molar from Mencal-9, assigned to the oldest recorded presence of Hippopotamus on the Iberian Peninsula (Arribas, pers.comm.);(k) incisor fragment determined as Hippopotamus amphibius found in the Sima de las Palomas de Cabezo Gordo, assigned to one of the last recorded presences of hippopotamuses on the Iberian Peninsula (Walter, pers.comm.);(l) cut mark reported on a bone of H. antiquus at the locality of Fuente Nueva (modified from[20]); (m) femur of H. antiquus with marks of exploitation by hyenas found at Fuente Nueva (modified from[20]); and (n) partial skeleton of H. antiquus found at Cal Guardiola.Scale bar (a-i) 3 cm, (j-k) 1 cm and (n) 10 cm.