The Science of Nature (2018) 105:63 httpsJ/doi.org/ 10.1007/ s001 14-018-1586-5

ORIGINAL PAPER CrossMark

Did the Romans introduce the Egyptian ( ichneumon) into the ?

4 5 Cleia Detry 1 E> . João Luís Cardoso 1,2,3 . Javier Heras Mora • Macarena Bustamante-Álvarez • Ana Maria Silva 1,6 . João Pimenta 1,7 • Isabel Fernandes 8 • Carlos Fernandes 9

Received : 19 June 2018 / Revised: 23 August 2018 / Accepted : 19 September 2018 © Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract New finds ofbones ofthe Egyptian Mongoose (Herpestes ichneumon), one from and one from , were directly 14C dated to the first century AD. While the Portuguese specimen was found without connection to the Chalcolithic occupation ofthe Pedra Furada cave where it was recovered, the Spanish fmd, collected in the city of Mérida, comes from a ritual pit that also contained three human and 40 burials. The finds rep0l1ed here show that the Egyptian mongoose, contrary to the traditional and predominant view, did not first atTive in the lbetian Peninsula during the Muslim occupation of lberia. lnstead, our findings are consistent with the hypothesis that the species was first introduced by the Romans, or at least sometime during the Roman occupation of . Therefore, rad iocarbon dating of new archaeological finds of bones of the Egyptian Mongoose (Herpestes ichneumon) in the lberian Peninsula pusb back the confimled presence ofthe species in tlle region by approximately eight centuries, as the previously oldest dated record is from the ninth century. With these new dates, there are now a total offour 14C dated specimens of Egyptian from the lberian Peninsula, and ali ofthese dates fali within the last 2000 years. Tms offers support for the hypothesis that the presence ofthe species in lberia is due to hi storical introductions and is at odds with a scenario of natural sweepstake dispersai across the Straits of Gibraltar in the Late (126,000- 11 ,700 years ago), recently proposed based on genetic data.

Keywords Egyptian mongoose . Herpestes ichneumon · lberia . Roman period

Communicated by: Sven ThaDe Introduction

18:1 Cleia DetIy The origin of the lberian population of the Egyptian mon­ [email protected] goose (Herpestes ichneumon) has been much debated. Zooarchaeology can play an important role in c1arif)ring when UNlARQ- Ceno'o de Arqueologia da Universidade de Lisboa, and how this species reached the Iberian Peninsula. Faculdade de Letras da Uni versidade de Lisboa, Lisbon, POlt ugal Today, tms matnmalian mesocarnivore is found mainly in 2 Universidade Aberta, Li sbon, POIt\lgal and southwest Asia (Do Linh San et a!. 2016). ln ICArEHB, Faculdade das Ciências Humanas e Sociais, Uni versidade Europe, its distribution is restricted to the southwestem pat1 do Algarve, Faro, POItugal of the Iberian Peninsula (Delibes 1982; Balmori and 4 Junta de Exo'emadura, Extremadura, Spain Carbonell 2012; Fig. 1). Deprutamento de PrehistOli a y Arqueología. Facultad de Filosofia y The distribution of the Egyptian mongoose in the lberian Leo'as, Uni versidad de Granada, Granada, Spain Peninsula may have experienced expansions and contractions, Laboratólio de Préhistória, CIAS- Cenl1'o de Investigação em with the species now undergoing a period of expansion and Ano'opologia e Saúde, Depaltamento Ciências da Vida, Uni versidade recolonization (BOtTalho et a!. 1996; Balmori and Carbonell de Coimbra, Coimbra, Portugal 2012; Barros et a!. 20 I 5). The fact that the mongoose inhabits Município de Vila Franca de Xira, Vila Franca de Xira, POItugal Mediterranean forest and SIUl.lblands and prefers dense vege­ Municí pio de Palmela, Palmela, POItugal tation cover, coupled with the reduction ofthese biotopes due CE3C- Centre for Ecology, Evolution and Environmental Changes, to agricultural expansion in the twentieth centuty, may explain Faculdade de Ciências da Uni versidade de Lisboa, Lisbon, POIt ugal the shrinkage ofthe species' range during that period (Delibes

~ Springer Published online: I I October 20 18 63 Page 2 of 13 Sei Nat (2018) 105:63

1982; Borralho et a!. 1996). Throughout much of the last from Nerja Cave in southern Spain, radioearbon dated century, the mongoose was restricted to the south of the to the twelfth eentury AD (Riquelme-Cantal et a\. Tagus , whereas in the nineteenth century, the lberian 2008), and another from Muge in central Portugal, range was more extensive and included some of the northern­ 14C dated to the ninth eentury AD. These were eon­ most regions of the peninsula, such as Galicia and Asturias sistent with the hypothesis of introduction(s) during (Delibes 1982). It is possible that before the nineteenth centu­ the Muslim oecupation of the lberian Peninsula ry, with lower human pressure, the Egyptian mongoose could (Detry et a\. 20 I I). More reeent!y, remains of the have been distributed across the Iberian Península. However, mongoose have been found welI to the north of its there are probably factors, such as lower temperatures (BalTOS present-day range, in G ijón, northern Spain, and dated et a!. 2015), that limit the spread and persistence of the species to the sixth eentury AD (L1orente-Rodríguez et a\. in the north, and the Pyrenees apparent1y act as a banier to 2015) (Fig. I). dispersai outside the peninsula. Here, we report additional remains of Egyptian 1110ngoose While the Egyptian mongoose is known from the from Portugal and Spain, two ofwhieh were 14C dated to the Late Pleistocene and Holocene of , it is Roman period (Table I; Fig. I). The Spanish set of remains absent in the European Pleistocene fo ssil record (Fig. 2) was dated to the first century AD and found together (Kurtén 1968; Dobson 1998). Accordingly, its presence with 40 in a ritual eontext in Mérida (Calle in the lberian Peninsula is generally considered the re­ Almendralejo), western Spain. The radioearbon-dated bone sult of historical introduction from north Africa (Dobson fr0111 Portugal (Fig. 3), also dated to the first centUJy AD, 1998), and traditionall y linked to the Arab conquest of was found intrusive (i.e. not eontemporaneous with the pre­ lberia (Delibes 1982). historie human oeeupation) in a Chalcolithie context in the Before this study, three archaeologieal finds of Cave of Pedra Furada in Vila Franca de Xira. The three re­ Egyptian mongoose from Iberia were published. One maining bones (Fig. 4) were found in lslamie eontexts in the

Fig . 1 Map showing the geographic range (grey colour) of the Egypti an mongoose (H el]Jestes ichneumon) in the Tberian Peninsula until 20 I O (Barros et aI. 2015 ; Balmori and CaI'bonell 2012). Black dots and stars, respectively, indicate the location of previously published and new finds (thi s study) of [-J. ichneumon. I- Fáblica de Tabacos, Gijón, Spain; 2- Moita do Sebastião, Muge, POItugal; 3- Cabeço da Amoreira, Muge, POItugal; 4- Pedra Furada I Cave, Vila Franca de Xira, Portugal; 5- Calle (street) Almendralejo, 4 1, Mérida, Spain; 6- Ponta da Passadeira, Barreiro, POItugal; 7- Palmela Castle, Palmela, POItugal; 8- Nelja Cave, Málaga, Spain. Map done by André Pereira

~ Springer oV1, z ~ 'N o ~ o V> 0\ w

Table 1 Archaeological finds of Egyptian mongoose (Herpestes ichneulI1 on) repOlted fi'om Europe, Dates are calibrated using Calib Rev, 7,0.4 with the curve Intcal13 (Reimer et aI. 2013)

Sample Counoy, region Bones Contex t Chro nology ofthe Date ofthe bones Lab Code 13C 14C Date Calibrated date Reference archaeological site (2u)

Tabaquera Spain, Gijon 28 bones Contemporaneous Late Antiquity 6th Cent. AD * :I: Not dated * Llorente et a I. Late Antiquity 20 15 Muge (Moita do POIt ugal, SalvatelTa Cranium lntrusive? Mesoli thic :I: Not dated * Deo'y et aI. 20 11 Sebasti ão) de Magos Muge (Cabeço POItugal , Salvaterra Ulna, Pelvis Ino'usive Mesoli thic 9th Cent. AD WK-26799 11 68 ± 30 772- 903 Deoy et aI. 20 11 da Amoreira) de Magos Islamic peri od 9 18- 965 cal AD Pedra Furada Cave POItugal, Vila Franca Ulna lntrusive Chalcolithic and 1st Cent. AD Beta-4673 10 - 19,7 2030 ± 30 154- 139 cal BC This 3I1icle, de Xira Bronze Age Roman peri od 11 3 cal BC - Fig, 3 52 cal AD Calle Almendralejo Spain, Méri da Cranium, Femur, Contemporaneous Roman 1st Cent. AD Beta-4673 I I - 16,9 2020 ± 30 104 cal BC - 57 This alticle, Tibia Roman period cal AD Fig, 2 Ponta da Passadeira Ponugal, BalTeiro Mandible In01Jsive? Neolithic * :I: Not dated * Soares 20 13 Palmela Castle POIt ugal, Humerus, Femur Contemporaneous? Medieval 8th/9th- 10th Cent. :I: * Not dated * This an icle Palmela and Tibia AD, Fig, 4 Islamic Period Nelj a- 1734 Spain , Málaga Cranium Intrusive Chalcolithic 12th Cent. AD Ua-32892 - 19,9885 ± 40 1033- 1224 Rique1me-Cantal Islamic period 1239- 1249 et aI. 2008 cal AD Sant' Antioco Italy, Sardinia Distal humerus Contemporaneous Punic 4th- 5th Cent. BC :I: * Not dated Campanela and Punic Wilkens 2004

\O" '"rtl w I ~ ~'" S' I~ aco !I 63 Page 4 of 13 Sei Nat (201 8) 105:63

Fig. 2 Bones ofEgyptian mongoose (He/yJestes ichneumon) fOllnd in the fllneraty pit at Call e Almendralejo, Mérida (Spain) (on the left), compared with the bones ofthe LARC­ DGPC reference collection (right). Photographs: José Paulo Ruas

o 1 2 3cm Iaeeeea! lweeew4 medieval site ofthe Castle ofPalmela, Portugal, and dated via Archaeological remains of the Egyptian the associated cultural remains. mongoose in Europe The finds discussed here are therefore the oldest radiocarbon-dated remains of Egyptian mongoose described lt has been argued (Masseti 2009) that the earliest fmd ofthe so far for the Ibelian Peninsula. Egyptian l110ngoose in Eurape is a bone recovered fram a

~ Springer Sei Nat (2018) 105:63 Page 5 of 13 63

Fig.3 One ulna identified as Egyptian mongoose (Hel/Jestes ichl1eumon) from Pedra Furada Cave (pOItugal) (on the left), compared with the bones of the LARC-DGPC reference collection (right), radiocarbon dated to the first centuly AD. Phot06~'aphs: José Paulo Ruas

o 2 3cm ! I

Punic cistem dated to the 5th-4th centUly BC in Sant' Antioco ofthe species, including 28 bones tl1at are very well preserved lsland, Sardinia (Campanella and Wilkens 2004). However, due to the low-oxygen content of the soil. They also do not this is a debatable asset1ion, given the possibility of strati­ show any signs of having been eaten or prepared in any way graphic uncertainty (e.g. Riquelme-Cantal et aI. 2008). (Llorente-Rodríguez et aI. 2015). The stratigraphic unit in Masseti (2009) convincingly makes the case that the bone which they were found also contained an abundance of pet probably belonged to a tamed or domestic kept for animais such as cats and dogs, and tlús raises the possibility controlling . The fact that the bone was found in a that the mongoose was also a pet (Llorente-Rodríguez et aI. closed deposit makes it likely that it was intentionally 2015). Gijón is not within the current distribution range ofthe deposited. species, but might have been in the past, with several authors The second published find from Europe, and the &st from the reportillg its presence in northem Spain. Gijón was also an lberian Peninsula, was unearthed in Netja Cave (Riquelme­ impotiant port in Late Antiquity (Fernandez-Ochoa et aI. Cantai et aI. 2008). Only one complete cranium was fOUlld , but 2015). its good preservation facilitated its identification to species. The There is also an undated isolated l11andible found at the layer from which it was recovered was cultnrally dated to the Neolithic open-air site of Ponta da Passadeira (Barreiro, Bronze Age, but a direct radiocarbon date on the specinlen Portngal) (Soat'es 2013). It is possible that this specimen is showed that it in fact belonged to a much later period, the twelfth intrusive since it came from an unsealed context in this centnry AD. Nerja cave was occupied intennittently during the open-air site. medieval period, including the lslamic one. New evidence for the presence of tlle Egyptian mongoose in More recently, in the Mesolithic shell middens of Muge, Roman lberia (Vila FratlCa de Xira and Métida) and Medieval we found four bones of H. ichneumon: one cranium in Moita Portugal (Palmela) is described in the "Results" section. do Sebastião, and one pelvis and two complete right ulnae in Table I presents a summary of the archaeological remains Cabeço da Amoreira (Detty et aI. 20 11). One ofthe ulnae was of H. ichneumon in Emope and see Detry et aI. (2011 ) for a radiocarbon dated to the ninth century AD, making it contem­ briefreview ofthe paleontological and archaeological records poraty with the Ul11ayyad Emirate of Cordoba, in the eat'ly of the species from elsewhere. phase of the lslamic occupation of lberia during medieval times. The latest published find of H. ichneumon in Europe comes Methods from a site in Gijón (La Fábrica de Tabacos), a location out­ side the current distribution range, and has been dated to the The mongoose remains repOlied in tlús stndy were found in sixth or seventh centnry AD (Llorente-Rodríguez et aI. 2015). general zooarchaeological surveys at the three archaeological It is the most complete skeleton among the Emopean rel11ains sites. Of the bones collected, those of smal1 camivores were

~ Springer 63 Page 6 of 13 Sei Nat (201 8) 105:63

Fig. 4 Bones identified as Eb'Y ptian mongoose (HellJes tes ichneumon) frum Pab11ela castle (portugal) (on the left), compared with the bones ofthe LARC­ DGPC reference collection (right). The archaeological context was dated to lhe 8th- 10th centuly AD. Photographs: José Paulo Ruas

o 2 3cm ! ! compared with the reference collection of the Archaeosciences measured separately in an isotope ratio mass spectrometer LaboratOlY of Lisbon (Direcção Geral do Pau;mónio Cultural). (IRMS). The 14C dates were then calibrated using Calib REV Ofthe bones identified as belonging to H. ichneumon, a sample 7.0.4 and the lntcal13 curve (Reirner et aI. 2013) (Table 1). of 2 g fi'om the Cave of Pedra Furada ulna and of the Mérida specirnen (tibia) were sent to Beta Analytic lnc. for radiocarbon dating. The protocol involved a pre-treaul1ent of the bone sam­ pies with cold HCL to eliminate the mineral fraction and extract Resu lts the collagen, which is the portion needed for radiocarbon dating. To ensure removal of secondaJ.Y organic acids, the collagen was The remains studied here come from three different washed with an alkali solution. The samples were then convelted sites: a ritual pit in Emerita Augusta (modem Mérida), into graphite for accelerator mass specu'ometly (AMS) analysis the Roman capital of Lusitania; a cave with prehistoric and the ratios of 14C, 13C and 12C were measured. We used AMS occupation c\ose to the Tagus estuary (Gruta da Pedra because it is the method of choice for radiocarbon measurements Furada); and the medieval castle of Palmela, in the due to its accuracy and requiring a much smaller saJ.l1ple than Setúbal Peninsula between the Tagus and Sado estuaries radiomeu'ic techniques (Wood 2015). The d 13C values were (Fig. 1).

~ Springer Sei Nat (2018) 105 :63 Page 7 of 13 63

Emerita Augusta (Calle Almendralejo) more recent occupation, as the chicken is known in Portugal mainly since the lron Age (Davis 2007). Between 2005 and 2007, emergency excavations were under­ The Egyptian mongoose ulna has an unfused olecranon, taken at number 41 , Calle Almendralejo, near the nOlth Gate indicating ajuvenile individual. The ulna was 14C dated yield­ of Emerita Augusta. At this location, a large Roman dump ing a 20" interval of 154 cal BC- 52 cal AD (Table 1), and was uncovered, containing remains dating c10se to the foun­ hence its presence is unconnected with the hUma.Il occup ation dation of the city, at the end of first century AD to the fifth ofthe cave. During the long interval between the Bronze Age century AD (Bustamante-Álvarez 2013; Heras Mora et aI. and modem times, the cave was unoccupied and would natu­ 2011 , 2017). ln the middle of the site, a large pit was found rally have been used by wild animaIs such as the mongoose. with evidence that indicate its ritual character, among them, The recovered specimen, in contrast to the remains from ceramics with a high symbolic content or fine tableware des­ Emerita Augusta, is intmsive and most certainly belonged to tined for symposia (Heras Mora et aI. in press; Heras Mora a wild animal. and Olmedo 2010). The H. ichneumon remains found at the site inc1ude a cra­ Palmela Castle nium (posterior part without maxilla), a right pelvis, a right femUf and one complete left tibia (Fig. 2). All the long bones Palmela castle in the town of Palmela, 40 km southeast of had completely fused epiphyses, ind icating that they belonged Lisbon, is an important fortress- occupied since the earli est to an adult animal. The tibia was directly radiocarbon dated to Muslim presence in Iberia- the end ofthe eighth centuly/fIrst 104 cal BC- 57 cal AD (Table 1). The remains were discov­ half of the ninth century AD (Fernandes 2004). ered in a particular archaeological context, with three human ln one of the excavated gallelies, in the oldest deposits, skeletons at the bottom ofthe pit covered by ca. 40 dog skel­ three very well-preserved bones- a left humems, a left pelvis etons and the mongoose bones. This assemblage had been and a right femur- were recovered (Fig. 4). The context is sealed with a layer of funerary banquet terra sigillata dated, through the materiais recovered, between the 8th/9th sudgalica (Bustamante-Álvarez 2009), indicating a date for and the 10th centuries AD. The mongoose remains could the pit contents of ca. 50- 60 AD. The 14C date is compatible not have been introduced before the constmction ofthe castle, with this supposition. The fact tllat the context was sealed and they are possibly coeval with the first Islamic occupation, made it less likely that the remains could have been intmsive. when the castle was built. The mongoose bones may have belonged to the same an­ imaI and, as in the case ofthe specimen from Gijón (Llorente­ Radiocarbon dating RodIiguez et aI. 2015), the fact that it was buried with dogs, many of which were small lap-dogs (Pires et aI. 2017), sug­ Direct radiocarbon dating is particularly useful in the case of gests that it could also have been a pet. burrowing animaIs, since this behaviour increases the likeli­ hood that their remains are intmsive into archaeological layers. For the Egyptian mongoose, this concem is reinforced Gruta da Pedra Furada by remains that yield ed radiocarbon dates l11uch younger than the archaeological layers in which they were recovered Pedra Furada are two caves located on a slope near Vi la (Riquelme Cantai et aI. Riquelme-Cantal et aI. 2008; Delly Franca de Xira, a town 40 km northeast of Lisbon. ln the first et aI. 2011 ). cave (Gruta da Pedra Furada 1), among other faunal remains, Two of the bones described here, one tibia from Emerita we found a left ulna of Egyptian mongoose (Fig. 3). The cave Augusta and one ulna from Gmta da Pedra Furada, were sent had an intelmittent late Chalcolithic and Bronze Age occupa­ in 20 17 to Beta Analyti c lnc. and both produced dates con­ tion and was also used in the early modem period (fifteenth tel11poraneous with the beginning of the Romanization of centl.lly AD). ln prehistoric times, the site was mainly used for Hispania after its conqu est. The overlap between tl1e two es­ burials and the skeletons of some 34 or more humans have timated dates (Beta-46731 Oand Beta-4673 1 1) was a complete been found. Only one human bone was direct1y radiocarbon surprise since the two bones come from vely different sites dated, and the estimated date fell in the 4tll/3rd millennium hundreds of kil ometres apart. The different 13C values con­ BC (Silva et aI. 2014), thus compatible with a Chalcolithic firmed that the bones were from different individuais with occupation. No Roman remains were found in the cave. lt different diets. This mies out the possibility that the similarity was excavated in the 1950s and the faunal remains were re­ ofthe two estimates could be due to sal11ple l11ixing or cross­ cently studied by Cardoso and Detry (see Silva et aI. 2014). contamination in the laboratory before analysis. Among the remains were domestic species such as cattle, The remains from Palmela Castle were not 14C dated be­ sheep, goat and pig. Most are interpreted as burial offerings cause they could not be older than the end of the eighth cen­ as was then customaty. A few bones of chicken may indicate a tury AD or the first half of the nintl1 centuly, when the castle

%l Springer 63 Page 8 of 13 Sei Nat (2018) 105:63 was built, and thus cannot set an earlier date for the arrival of derived from westward dispersai from the Middle East across the Egyptian mongoose in the Iberian Peninsula than that al­ Southem Europe, given the CUlTent absence of prehistoric re­ ready established by the two bones radiocarbon dated in this mains of the species in Europe. study (first centUly AD). Since the opening of the , sea levei in the strait was lowest 30 ka and 26- 17 ka ago, when it was about 125 m, and at most 150 m, below present sea levei Discussion (Rohling et aI. 2014). On the west side of the strait, the Camarinal Sill, the shallowest sill separating the There is currently no evidence that the Egyptian mongoose MeditelTanean Sea from the Atlantic Ocean, has a present has ever been tmly domesticated, in the sense of selective depfu of 284 m (Blanc 2002). Paleogeographic reconstmc­ breeding across multiple generations leading to permanent tions indicate that the minimum width of the strait in the genetic modification (Driscoll et aI. 2009). Nevertheless, it is Pleistocene would have been 8- 10 km (Shackleton et aI. knOWl1 that the species can be tamed and bred as a household 1984; Collina-Girard 2001 ; Gutscher 2005 ; Gracia et aI. pet (Harrison 1968; Osbom and Helmy 1980; Ben-Yaacov 2008). To the west of the Camarinal Si II, where the strait and Yom-Tov 1983; Evans 2016) and has historically been widens, some presently submerged small islands and islets, used for control (Faure and Kitchener 2009; Masseti including the Spm1el Bank, would have appeared whenever 2009). This latter purpose has been suggested as the main sea levei fell 100- 130 m below its present levei (CoUina­ reason for the introduction of the Egyptian mongoose in the Girard 2001 ), while remaining separated by deep channels lberian Peninsula (Faure and Kitchener 2009; Masseti 2009; (Blanc 2002). The use of these islands as stepping stones for DetJy et aI. 2011 ). As noted by Masseti (2009), the reality of dispersai across the strait would have required at least two ancient introductions ofthe species to Europe is attested by the hops, one of them involving a distance over water of no less bone from Sant' Antioco Island. Ancient introductions to than 6- 7 km (Collina-Girard 2001 ; Gracia et aI. 2008; Sicily and the ltalian Peninsula in late Antiquity or medieval Linstãdter et aI. 2012). Moreover, during cold periods such times also li.kely explain the early presence ofanother African as the last glacial maximum (LGM, 26.5- 19 ka ago; Clark , the crested porcupine Hystrix cristata, probably for et aI. 2009), marine surface currents flowing westwards phamlaceutical purposes (Masseti et aI. 2010). (Gibert et aI. 2003) would tend to carry dispersing animais, Recently, based on the results of a genetic study, it has been either swimming or on natural rafts, into the Atlantic and not postulated that the Egyptian mongoose entered the lberian across the strait. Vegetation rafts, unli.ke in the tropics, are Peninsula by natural sweepstake dispersai across t11e Strait of not a typical feature of the Mediterranean today, and proba­ Gibraltar sometinle dming fue Middle to Late Pleistocene cyclic bly were not during the Pleistocene (Broodbank 2006). lowering of sea levei (Gaubel1 et aI. 20 11 ; Gaubelt 2016). Accordingly, the palaeontological and zoogeographical data The Strait of Gibraltar has been open since the end of the indicate that the Strait of Gibraltar has been an effective Messinian salinity crisis at the Miocene-Pliocene transition, barrier to a large number of land animal species (O 'Regan ca. 5.3 million years ago (Krijgsman et aI. 1999; Blanc 2008), and of those that have crossed several result from 2002; Loget and Van Den Driessche 2006). lmportantly, while human introductions (Dobson 1998; Cosson et aI. 2005; the earliest fossils assigned to Herpestes are from the Middle Modolo et aI. 2005; Lalis et aI. 2016). Palaeontological ev­ Miocene (ca. 14.5 Ma; Tugen Hills, Kenya; Morales and idence of the native presence of a species on both sides of Pickford 2008) and Late Miocene (9.5- 7 Ma; Siwaliks, the Strait in the Plio-Pleistocene does not necessarily imply Pakistan; Barry 1983) (ca. 7 Ma; Toros-Menalla, Chad; sea crossings, since it could also be due to migration around Peigné et aI. 2008), H. ichneumon is first known from the both shores of the MeditelTanean (O'Regan 2008). Among mid-Pliocene Upper Laetolil Beds (3.85- 3.63 Ma; Deino those species that could plausibly have crossed the Strait are 2012) at Laetoli, northem Tanzania (Petter 1987; Werdelin Im·ge with good swimming abilities, such as , and Dehghani 2011 ). The Egyptian mongoose has also been deer, and possibly boar and hippopotamus (Dobson 1998; reported from the early Middle Pleistocene of Asbole, O'Regan 2008; Geraads 2010; Palombo 2014; Femandez (Geraads et aI. 2004) and from the Early et aI. 2015; Soria-Boix et aI. 2017). lt is therefore clem· that Pleistocene site of Ubeidiya, Israel (Belmaker 20 I O). the Strait has acted as a stJ·ong biogeographic filter, allowing ln northwestem Africa, the species is found from the the dispersai of a limited number of species (Fen·àndez• Middle Pleistocene onwards (Geraads 2008; Geraads et aI. Caiíadell et aI. 2014; Gibert et aI. 2016). Besides the 2010; Raynal et aI. 2010). Thus, the species apparently did Egyptian mongoose, the only other Afrotropical camivore not exist, or at least did not occm in north Africa, before the in the lberian Peninsula, the common (Genetta opening of the Strait of Gibraltar, mling out the scenario of genetta), is widely considered to have been introduced from dly-Iand crossing of the MeditelTanean. However, it seems north Africa (Morales-Muruz 1994; Dobson 1998; Gaubert unlikely that the Iberian population of H. ichneumon is et aI. 2011 ; Gaubert 2016).

%l Springer Sei Nat (2018) 105:63 Page 9 of 13 63

Even archaeological evidence for hWllans throughout the Pleistocene, it is striking that none of the dated remains of Pleistocene has been generally interpreted as indicating that lberian H. ichneul110n are prebistoric (Table 1). trans-Gibraltar contacts only occurred at the end ofthe Upper Some of the remains (Emerila Augusta and Gijón) were Palaeolithic, about 12,000 years ago (Straus 2001; Erlandson contemporaneous with the archaeological contexts in which 200 I ; Garcea 2004; Derricourt 2005; Broodbank 2006; they were found and may represent pets that were intention­ Carbonell and Rodríguez 2006; Alcaraz Castano 2007; van ally buried. lt might be argued that the age and context ofthese der Made 2011 ; Linstadter et aI. 2012; Garcia et aI. 2013; specimens, since they could have been pets, do not necessarily Croitor 2018; Muttoni et aI. 2018), presumably by boat since on the age and origin ofthe wild population in the lberian by this time, sea levei was similar to the present (Rohling et aI. Peninsula. However, this caveat does not apply to the remains 2014). Only recently, evidence has been accumulating to sug­ from Nerja, Muge and Gmta da Pedra Furada, which were ali gest that the lberian Acheulean, a type ofhuman litbic indus­ shown by direct radiocarbon dating to be intmsive. These t1y, may have originated in n011h Africa through contact across intmsive remains are likely the result of burrowing behaviour the Strait of Gibraltar during the early Middle Pleistocene (Palomares and De\ibes 1993), and therefore most probably (Santonja and Pérez-González 2010; Jiménez-Arenas et aI. belonged to individuais living in the wild. 2011 ; Sharon 2011 ; Sharon and Barsky 2016; Santonja et aI. The dated H. ichneumon remains from Europe reported so 2016; Álvarez-Posada et aI. 2017; but see also Rolland 2013 far seem to suggest an interesting pattem, which may apply to and Walker et aI. 2013). Nevertheless, the apparent overall other blmowing animais that could have been tamed, kept as rarity of trans-Gibraltar human movements in the pets or venerated in some way. On one hand, there are spec­ Pleistocene reinforces the idea that the Strait is a strong barrier imens, often isolated bones, shown to be intmsive by radio­ to dispersai of land mammals. carbon dating and recovered from archaeological contexts Gaubert et aI. (2011 ) proposed the hypothesis that the (much) older than the age of the remains, which probably Egyptian mongoose reached the lberian Peninsula via a belonged to wild individuais that died in their burrows. On sweepstake crossing ofthe Strait ofGibraltar sometime during the other hand, there are finds, in some cases of several bones, the Middle to Late Pleistocene, based mainly on the following coming from sealed loci and ritual burials, and dated as con­ results of their genetic study: (i) an observed high mitochon­ temporaneous with the contexts in which they were found, drial DNA (mtDNA) differentiation between population sam­ which likely represent pet or tame animais. pies from lberia and northwest Africa and (ii) the levei of Here, we present two assemblages representing these two mtDNA diversity in the lberian sample, which was considered depositional types, and both dated to the first century AD. high by the authors. However, sample sizes were unbalanced They unambiguously prove the presence ofthe Egypti an mon­ between lberia and northwest Africa, being very small for goose in !beria during the Roman period and represent the (/1 = 3) and AIgeria (n = 7). lt is well known that oldest record of the species in the lberian Perunsula. comparisons of mtDNA diversity between populations or re­ Our f1l1dings are thus consistent with the hypothesis that gions are highly sensitive to undersampling (Goodall­ the species was first introduced by the Romans, or at least Copestake et aI. 2012; Phillips et aI. 2015). Small and unbal­ sometime dming the Roman occupation of Hispania. lt is anced sample sizes can also lead to biased estimates of genetic well known that during the Roman Republic and Empire, differentiation due to sampling erro r on allele frequencies there was a constant movement of people, animais, goods (Holsinger and Weir 2009; Caujapé-Castells 20 I O), and tbis and ideas ali over the Mediterranean basin, linking north bias is particularly likely with a single locus such as mtDNA. Africa and the Levant with Southem Europe (Leitch 2013; lnsufficient sampling in n011hwest Africa may have missed Scheidel 2014; Colominas and Edwards 20 I 7; Pires et aI. haplotypes shared with the lberian Peninsula if they are cur­ 2017). ln Emerita Augusta, evidence of a north African rently infrequent in the fomler region. In such cases, it has connection is illustrated by the presence of Apican Red been shown that inadequate sampling can strongly influence Slip Ware from the province of Byzacena () until results and interpretations on the origin of extralimital popu­ the sixth centuIy AD, and also the trade of oil amphorae lations (Cristescu et aI. 2001 ; Hanfling et aI. 2002; Muirhead ofthe sarne origin (Vázquez de la Cueva 1985). We can also et aI. 2008; Johnson et aI. 2011 ). observe a religious connection since the funenuy monument The new dated finds repo11ed here, taken together with where the mongoose was found seems to have a n011h-Afi"ican dates from previous studies (Riquelme-Cantal et aI. 2008; influence (Prados 2008), such as the cupae and tower mauso­ Det1y et aI. 2011 ; LlOl'ente-Rodríguez et aI. 2015), strongly leums (Bustamante-Álvarez and Olmedo 2012). The ritual support the hypothesis that the presence ofthe Egyptian mon­ sacrifice of dogs, which is probably the case at Calle goose in the lberian Peninsula is due to histOlical introduc­ Almedralejo, is also related to Punic-Hispanic influences and tions. As previously remarked by Detly et aI. (2011 ), if as of north-African tradition, as it was interpreted for the necrop­ proposed by Gaubert et aI. (2011 ) the species had arrived in olis of Gadir (modem Cádiz, Spain) (Niveau and Marinas the lberian Peninsula some time in the Middle to Late 2006; Heras Mora et aI. in press).

~ Springer 63 Page 10 of 13 Sei Nat (201 8) 105:63

As discussed above, the association with dogs suggests that Álvarez-Posada C, Parés 1M, Sala R, Vi seras C, Pla-Pueyo S (201 7) New the Egyptian mongoose bones that were recovered from the magnetostratigraphic evidence for the age of Acheul ean tools at the archaeo-palaeontological site "Solana dei Zamborino"(Guadix­ sarne site may have belonged to an intentionally buried pet. ln Baza Basin, S Spain). Sci Rep 7(1): 13495. https://doi.orgIIO. this context, it is worth remembering that in ancient , the I 038/s4 1598-0 17-14024-5 mongoose c1ear1y had some religious and emotional signifi­ Balmori A, Carbonell R (201 2) Expansion and di stribution of the cance: it was represented in paintings, collars, statues, burials Egyptian mongoose (Herpestes ichneulI1 on) in th e lberi an Pe nins ul a . G a le mys 24: 83- 8 5. https ://doi .o rgIl0.7325/ and was even mummified (Hinton and Dunn 1967; Nicholson Galemys20 12.N08 et aI. 2015; Evans 2016). Barras T, Carva lho J, Pereira MJR, Ferreira JP, Fonseca C (2 01 5) Foll owing the trail : factors underlying the sudden expansion of the egypti an mongoose (Hei p estes ichneumon) in POlt uga l. PLoS One I 0(8):eOI33768. https://doi.orglI0.1371/joumal.pone.0 133768 Conclusion Barry JC (1983) Herpestes (Vivenidae, Cami vora) fi'om the Miocene of Pakistan. J Paleontol 57(1): 150- 156 Belmaker M (2 0 I O) Ea rl y Pleistocene faunal connections between Afli ca This report increases awareness ofthe question ofthe origin of and Euras ia: an ecological perspecti ve. ln: Out of Afi'ica 1. Springer, the Egyptian mongoose in the lberian Peninsula and stimu­ Dordrecht, pp 183- 205 lates further research to shed more light on how and when this Ben-Yaacov R, Yom-Tov Y (1983) On the biolof,'Y ofthe Ef,'Y ptian mon­ mesocarnivore alTived in Iberia. For instance, it would be goose, He rpestes ic hne umo n, in Israel. Zeitschrift flir relevant to obtain radiocarbon dates for the remains in Sãugeti erkunde 48(1 ):34-45 Blanc PL (2002) The opening of the Plio-quaternary Gibraltar Strait: Table I not yet absolutely dated. assessing the size of a cataclysm. Geodin Acta 15(5- 6):303- 3 17. The findings presented here provi de the earliest record of https://do i.org/ 10.1080/09853111.2002.10510763 the species in the region and are consistent with a time offtrst Borralho R, Rego F, Palomares F, Hora A (1 996) The dishibution ofthe introduction during the Roman occupation of Hispania. Our Egyptian mongoose Herpestes ichneul1! on (L.) in Portu ga l. Mammal Rev 26: 1- 8. https: //doi.orgII0. IIII/j .1365-2907.1996. study therefore offers the first conc\usive evidence refuting the tbOOl43 .x traditional and dominant view that the first introduction oc­ Broodbank C (2 006) The o ri g ins a nd earl y development of cuned during the Muslim occupation of lberia. Moreover, the Mediterranean maritime activity. J Mediterr Archaeol 19(2): 199- new dated finds, taken together with dates from previous stud­ 230 ies (Riquelme-Cantal et aI. 2008; Detry et aI. 2011 ; Llorente­ Bustamante-Álvarez M (2 009) La terra sigill ata Gálica en Augusta Emerita (Mélida, Badajoz). Saguntum 41 : 149- 173 RodIiguez et aI. 2015), raise the possibility of multiple intro­ Bustamante-Álvarez M (201 3) La terra sigillata hispánica en Augusta ductions at different times fr0111 Classical Antiquity to the Emerita: estudio tipocronológico a partir de los vertederos dei Ear1y Middle Ages. suburbio nOIt e. Anejos de AEspA, LXV, Mérida The Egyptian mongoose bones from Emerita Augusta are Bustamante-Álvarez M, Olmedo A (2012) De las "cupae" emelitenses: nuevos datos estrati gráficos. ln Andreu J (ed) Las cupae hispanas notable because they come from an antlu'opic context, a ritual oligen, difusión, uso, tipolof,>Ía, UNED, pp 369- 392 pit that also contained three human and 40 dog burials, and Campanella L, Wilkens B (2004) Una mangusta egiziana ('Helpestes this leads us to believe that the mongoose bones belonged to a lchneul11 on') dali abitato fenicio di Sant Antioco. Riv Stud Fenici pet, which received the sarne ritual treatment. 32( I ):25-48 Carbonell E, Rodríguez XP (2 006) The first human settl ement of Mediten'anean Eurape. Comptes Rendus Palevol 5(1 - 2):29 1- 298. Acknowledgments We thank the Archaeoscience lab of the POItuguese https://doi.org/ IO. I O1 6/j.crpv.2005.12.002 Heritage directorate (LARC-DGPC) for use of their reference coll ec ti on Caujapé-Castells J (20 I O) General G ST and inflati on due to biased of modem mantinal skeletons, José Paul o Ruas for his excellent photo­ e inh'a- population sampling, and its consequences for the conselva­ graphs, and André Pereira fo r drafting the map. Finall y, we would also ti on of the Canarian Flora. Conselv Genet 11 (3):709- 720. https:// like to thank Simon Davis for revising the manu script and making doi.orgll 0.1007 /s I 0592-009-9842-z suggestions. Clark PU, Dyke AS, Shakun m, Carlson AE, Clark J, Wohlfalt h B, MitTovica JX, Hostetler SW, McCabe AM (2009) The last glacial Funding information UNIARQ (Center of Archaeology, Faculty of max imum. Science 325(5941 ):7 10 - 7 14. https: //doi.orgIl0. 11 26/ Letters of the University of Lisbon) funded the 14C dating, and FCT science. II 72873 (Fundação para a Ciência e Tecnologia) funded CD's post-doctorate Collina-Girard J (2001 ) L'Atl antide devant le détroit de Gibraltar? Mythe (grant: SFRHIBPD/ I 08326/20 15). CF was suppolted by FCT, within et géologie. C. R.Acad Sci Ser TI A Earth Planet Sci 333(4):233- 240. the project UIDIBW00329/20 13. https://doi.org/ IO. I O16 /S 125 1-8050(0 1)0 1629-9 Colominas L, Edwards CJ (20 17) Livestock h-ade dUling the early Roman period: fU'st c1ues from the trading post of Empúries (Catalonia). lnt J Osteoarchaeol 27(2): 167- 179. https://doi.org/ I 0.1002 /oa.2527 References Cosson JF, Hutterer R, Libois R, Sara M, Taberlet P, Vogel P (2005) Phylogeographical footprints ofthe strait ofGibmltar and quatemruy Alcaraz Castano M (2007) E I Ateri ense dei Norte de África y el c1imati c flu ctuations in the westem Med iten'anean: a case study with So lutrense peninsul ar:!. contactos transgibraltare l1 0s en el the greater white-tooth ed shrew, Crocidura ru ssula (Mammalia: Pleistoceno Superior? Munibe Antropologia-Arkeologia 58: 101 - Soricidae). Moi Ecol 14(4): 11 5 1- 11 62. https://doi .orglIO . IIII /j. 126 I 365-294X.2005.024 76.x

~ Springer Sei Nat (201 8) 105:63 Page 11 of 13 63

Clistescu ME, Hebelt PD, Witt ID, MacIsaac HJ, Grigorovich IA (200 I) Untell11assfeld (Germany) and Vallparadís (Spain). Quat Int 3 16:73- An in vasion histOly for Cercopagis pengoi based on mitochondrial 93. https://doi. orgIl O.1O 1 6/j .quaint.20 13.03.005 gene sequences. Limnol Oceanogr 46(2):224-229. https://doi.org/ Gallbelt P (201 6) Fate ofthe Mongooses and the Genet (Carni vora) in 10.4319/10.2001.46.2.0224 Medi ten·allean Euro pe: one native, ali invasive? ln: Angeli ci FM, Croitor R (2 01 8) Paleobiogeography of earl y humall dispersa I in westem problematic wildlife, Splinger, Cham, pp 295- 3 14. https:lldoi.orgl Euras ia: preliminalY results. Comptes Rendus Palevol 17(4-5):276- 10.1007/978-3 -319-22246-2_14, 201 6 286. https:lldoi.orgII O.1 O1 6/j.crpv.20 17.09.004 Gaubelt P, Machordom A, MOI·a les A, López-Bao IV, Verun G, Amin M, Dav is SJ (2007) The mammals and birds from the lron Age and Roman Barros T, Basuony M, Dj agoun C, Linh San E, Fonseca C, Geffen E, peri ods of Castro Marim, Algarve, POItuga l. Trabalhos do CIPA, Ozkult SO, Cruaud C, Couloux A, Palomares F (20 II ) Comparative 107. Instituto Português de Arqueologia, Lisboa phylogeography oftwo Afri can carnivorans presumably introduced Deino AL (2012) 40AIi39Ar datillg of bed I, Olduva i Gorge, Tanzania, into Europe: di sentangling natural versus human-mediated dispersaI and the chronology of earl y Pleistocene climate change. J Hum Evol acruss the su·ait of Gibraltar. J Biogeogr 38(2):341 - 358. https:lldoi. 63(2):251 - 273. https://doi.orgII 0. 101 6/j.jhevoI. 201 2.05 .004 orgll 0.llll /j. 13 65 -2699.2 01 0.02406.x Delibes M (1982) Notas sobre la distribución pasada y actu al deI Geraads D (2 008) Plio-Pleistocene of nOlthwestem Afi·i ca: a meloncill o Herpestes ichneul110n (L.) en la Peninsula Ibéri ca. short review. Comptes Rendus Palevol 7(8):59 1- 599. https:l/doi. Donana Acta Vertebrata 9:341 - 352 org/l 0.10 16/j .crpv.2008.09.00 8 Derricoult R (2005) Getting "out of Aii-ica": sea crossings, land cross ings Geraads D (20 I O) Biogeographic relati onships of Pliocene and and culture in the hominin migrations. J World Prehist 19(2): 11 9- Pl eistocene north-westem Afri can mal11mals. Quat Int 21 2(2): 159- 132. https:l/doi.orgl I0.1007/s 10963-006-9002-z 168. https:lldoi.org/ IO.1 O16 /j .quaint.2009 .06.002 Detly C, Bicho N, Femandes H, Femandes C (2 0 II ) The emirate of Geraads D, Alemseged Z, Reed D, Wynn J, Roman DC (2 004) The Córdoba (756- 929 AD) and the introduction of the Egypti all mon­ Pleistocene fauna (other than plimates) frum Asbole, lower Awash goose (Helpestes ichneu111 on) in Iberi a: the remains from Muge, Va lley, Ethiopia, and its environmental and biochronological impli­ Portu gal. J Archaeol Sci 38(12):35 18- 3523. htt ps:lldoi.orgIIO. cations. Geobios 37(6):697- 7 18. https:lldoi.orgII 0. 101 6/j.geobios. 10 16/j jas.20 11 .08.0 14 2003.05.011 Geraads D, Raynal Sbihi-Alaoui FZ (20 I O) Mammalian faunas frum Do Linh San E, Maddock AH, Gaubelt P, Palomares F (201 6) Herpestes JP, the Pliocene and Pleistocene of Casablanca (Morucco). Hist Biol ichneumon. The IUC red li st of threatened species 201 6: 22(1 - 3):275-285. https:lldoi.orgII 0. 10801089 12960903458011 e.T41 6 13A452072 11. doi: hrtps:lldo i.orglI 0.23 05/IUCN.UK. 201 6- I.RLTS.T41 613A452072 II .en. Accessed 29 May 201 8 Gibelt J, Gibel1 L, Igles ias A (2003) The Gibraltar Strait: a Pleistocene door of Eurupe? Hum Evol 18(3-4): 147- 160. https://doi.orgIIO. Dobson M (1998) Mammal distributions in the westem Meditemmean: 1007/BF02436283 the role of human intervention. Mammal Rev 28(2):77- 88. https:// GibeIt L, Scott GR, Scholz D, Budsky A, Ferrandez C, Ribot F, Marín doi. orgll 0.104 6/j . I 365-2907.1998. 00027.x RA, LerÍa M (201 6) Chronolob'Y fo r the Cueva VictOlia foss il site Dliscoll CA, Macdonald DW, O'Brien SJ (2009) Fro m wild animaIs to (SE Spain): evidence for earl y Pleistocene afi·o-Iberian dispersaIs. J domestic pets, an evolutionmy view of domesti cation. Proc Natl Hum Evol 90: 183- 197. https:l/doi.orgIl O.1O 1 6/j jhevol. 20 15 .08. Acad Sci 106(S upplement 1):997 1- 9978. https://doi. org/10.1073 / 002 pnas.090 1586 10 6 Goodall-Copestake WP, Tarling GA, Mlllphy EJ (201 2) On the compar­ Erl andson JM (200 I) The archaeolob'Y of aquatic adaptations: paradigms ison of population-I evel estimates of haplotype and nucleotide di­ fo r a new millennium. J Arch aeol Res 9(4):287- 350. https:lldoi.orgl versity: a case study using the gene cox I in animaIs. Heredity 10.10 23/A : 101 30627 12695 I 09( I ):50- 56. htt ps:l/doi. orgII 0.1038/hdy.201 2. 12 Evans L (201 6) Beasts and beli efs at Beni HasslU1: a preliminlllY repOlt. J Gracia FJ, Rodríguez-Vidal J, Cáceres LM, Belluomini G, Benavente J, Am Res Center Egypt 52:2 19- 229 http://lockwoodonlinejoumals. Alonso C (2008) Diapiric uplift ofan MIS 3 mllline deposit in SW com/index .php/jarce/alt iclelviewl78 Spain: implications for Late Pleistocene Sea leveI reconstruction and Faure E, Kitchener AC (2009) An archaeological and histOlical review of palaeogeography ofthe strait ofGibraltar. Quat Sci Rev 27(23- 24): the relationships between fe lids and people. Anthrozoos 22(3):22 1- 22 19- 223 1. https:l/doi. org/ 10. 101 6/j.quascirev.2008. 08.013 238. https:lldoi.orgII 0.2752/1753 03709X457577 Gutscher MA (2005) DesU·ucti on of Atl anti s by a great e3lthquake and Fernandes ICF (2004) O Castelo de Palmela. Do islâmico ao Cri stão, tsunami? A geological analysis of the Spartel Bank hypothesis. Câmara Municipal de Palmela, Palmela Geology 33(8):685- 688. https:lldoi.org/ 10. 11 30/G2 1597 AR.1 Fernández Ochoa C, GutiélTez Gonzá lez JA, Orejas Saco deI Va lle A Hiinfling B, Carvalho GR, BrandI R (2002) Mt-DNA sequences and (201 5) Gijón, entre la All tigüedad y la Edad Media. ln: Ochoa C possible invasion pathways of the Chinese mitten crab. Mar Ecol (ed) La fábrica de tabacos de Gijón: arqueologia e hi storia de un Prog Ser 238:307- 3 10. https:lldoi.orgII 0.3 354/meps238307 espacio milenári o, Ayuntamiento de Gijon, Gij on, pp 282- 295 Harrison DL (1968) The mammals of Arabi a. Ern st Benn Limited, http://hdl.handle.netII 065 1/39 195 London Fernandez P, Bouzoll ggar A, Collina-G irard J, COlllon M (201 5) The last Heras Mora FJ, Olmedo Gragera AB Pérez Maesu·o C (201 7) Dinámica OCC lllTence ofMegaceruides algericus Lyddekkel; 1890 (Mammalia, urbana en el Suburbio NOite de A ugusta Emerita. Síntesis di acrónica Cervidae) during the middle Holocene in the cave of Bi zmoune de las excavaciones en el llamado Corralón de los B lanes. ln: (Morocco, Essaouira region). Quat Int 374: 154- 167. https:lldoi. Mélida, Excav. Arqueol. 12 , vol. 2, Mérida orgll 0.1O 1 6/j.quaint.20 15.03. 034 Heras Mora FJ, Bustamante Álvarez M, Olmedo Gragera AB (2 0 II ) EI FelTàndez-Canadell C, Ribot F, Gibert L (2014) New fossil teeth of veltedero de suburbio nOite de Augusta Emerita. Refl ex ión sobre la Theropithecus oswaldi (Cercopithecoidea) fro m th e Earl y dinámica topográfi ca en el solar de la CI Almendralejo nO 41. ln: Pl eistocene at Cueva VictOlia (SE Spain). J Hum Evol 74:55- 66. Remolá JA, Acero, J (eds) La gesti ón de residuos urbanos en https:lldoi.org/ 10. 101 6/j.jhevoI. 2014.02.020 Hispania, Xavier Dupré Raventós (1956- 2006), in memori am, Garcea EA (2004) Crossing deselts and avoiding seas: Ateri an nOlth Mélida, pp. 345-4 18 African-European relations. J Anthro pol Res 60(1) :27- 53. https:ll Heras Mo ra F J, Bustama nte Á lva rez M , Detry C (in press) doi. org/l 0.10 86/jar.60.1 .3 63 100 7 Banquete y sacrificio ritual en Augusta Emerita (Mérida, Garcia J, Landeck G, MaItínez K, Carbonell E (201 3) Hominin dispersaIs Espana). Analizand o prácticas de posible ascendencia púnica. frum the Jara millo subchron in central and South-Westem Euro pe: ln: Proceedings X Coloquio Internacional deI Centro de

~ Springer 63 Page 12 of 13 Sei Nat (20 18) 105 :63

Estu dios Fenicios y púnicos. Mare Sacrum. Reli gión, cul tos y Nicholson PT, Ikram S, Mills S (201 5) The catacombs of Anu bis at nOlth rituales fenicios en el Mediterráneo. Homenaj e aI profesor Saqqm·a. Antiquity 89(345): 645- 661. https://doi.orgII0.15 184/aqy. José Marí a B lázquez (Cádi z, 13-1 5 december 201 7) 2014.53 Heras Mora J, Olmedo AB (2010) Dos nuevos monumentos funeralios en Niveau AM, Marinas VY (2006) Banquetes ri tuales en la necrópoli s Augusta Emelita. Prime r avance de los resul tados de la intervención púnica de Gadir. Gerión 24( I ):35- 64 de la calle A lmendralej o nO 4 1, Mérida (Badaj oz, Espalja). O'Regan HJ (2008) The Iberian Penin sula- corridor or cul-de-sac? Bollettino di Archeologia on line 1:45- 53 Mammalian faunal change and possible routes of dispersaI in the Hinton H, Dunn A (1 967) Mongooses. Their natu ra l hi stOIy and behav­ last 2 million years. Quat Sci Rev 27(23- 24):2136- 2 144. https://doi. iour. Uni versity ofCalifornia Press, Berkeley org/ I 0. 10 16/j .quascirev.2008.08.007 Holsinger K.E, Weir BS (2009) Genetics in geographicall y structured Osborn DJ, Helmy I (1980) The contemporary land mammals of E!''Y pt populati ons: defining, estimating and interpreting F ST. Nat Rev (including Sinai). Fieldiana Zoology 5: 1- 579 Genet 10(9):639- 650. https://doi.orgIl0.1038/nrg2611 Palomares F, Deli bes M ( 1993) Resting ecology and behaviour of Jiménez-Arenas 1M, Santonja M, Botella M, Palmqvist P (2011 ) The Egyptian mongooses (Herpestes ichneumon) in southwestern o ldest handaxes in Europe: fact or mtefact? I Archaeol Sci 38(1 2): Spain. J Zool 230(4):557- 566 3340- 3349. https://doi.org/IO. I O1 6/j.jas.20 11.07.020 Palombo MR (2014) Deconstructing mammal dispersaIs and faunal dy­ Johnson JR, Thomson RC, Micheletti SJ, Shaffer HB (20 II ) The origin nmnics in SW Europe during the quaternary. Quat Sci Rev 96:50- of tiger salamander (Ambystoma tigrinum) populations in 7 1. https://doi .orgIlO.1O 16 /j.quascirev.20 14.05 .0 13 California, Oregon, and Nevada: introductions or reli cts? Conselv Peigné S, de Bonis L, Mackaye HT, Likius A, Vignaud P, Brunet M Genet 12(2):355- 370. https:l/doi.orgIl0.1007/s I0592-010-0144-2 (2008) Late M iocene Carnivora from Chad : H erpestidae, Krij gsman W, Hilgen FJ, Raffi I, S ierro FJ, Wilson DS ( 1999) and small-sized . Comptes Rendus Palevol 7(8): Chronology, causes and progression ofthe Messinian salinity crisis. 499- 527. https://doi.orgIl0. 101 6/j.crpv.2008.10.002 Nature 400(6745):652- 655. ht1ps:l/doi.orgll 0.1038 /23231 Petter G (1987) Sma ll camivores (Viverridae, , ) from Kurten B ( 1968) Pleistocene mammals of Euro pe. We indenfeld & Laeto li . ln: Leakey M, Harris J (eds) Laetoli: a Pliocene site in Nicolson, nOlthem Tanzani a. Clarendon Press, Oxford , pp 194-234 Lalis A, Leblois R, Liefried S, Ouarour A, Reddy Beeravolu C, Michaux Phillips m, Gwiazdowski RA, Ashlock D, Hm1l1er R (201 5) An explo­ J, Hamani A, Denys C, N icolas V (201 6) New molecular data favour ration of sufficient sampling effOIt to describe intraspecifi c DNA an anthropogenic introduction of the wood mouse (Apodeml.ls bar"Code haploty pe di versity: examples fro m the ray-finned fi shes sy/vaticus) in North Africa. J Zool Syst Evol Res 54( I): 1- 12. (Chord ata: Actinoptelygii). DNA Bal"Codes 3(1):66- 73. https://doi. https:/ldoi.org/I 0. llll/jzs. 12 111 orglI 0.15 15/dna-20 15-0008 Linstiidter J, Eiwanger J, M ikdad A, Weniger GC (201 2) Human occu­ Pires AE, Deny C, Femandez-Rodriguez C, Arruda AM, De Grossi pation of N OIt hwest Africa: a review of middle Palaeolithic to Mazzorin I , Va lenzuela S, Ollivier M, Hanni C, Simões F, Ginja C Epipalaeolithic sites in Morocco. Quat Int 274: 158- 174. https:11 (201 7) Roman dogs from the Iberi an Peninsula and the Maghreb - a doi .org/ l 0.1O 16 /j.quaint.20 12.02.0 17 glimpse into their morphology and genetics. Quat Int47 1: 132- 146. Llorente-Rodríguez L, Nores-Quesada C, Gaubelt P. , Morales-Muniz, A. https:l/doi.org/lO.1 O1 6/j.quaint.20 17. 11 .044 (2 01 5). i,Nati vo o foráneo? E I meloncillo de Tabacalera y su Prados F (2008) Arquitectura púnica. ln: Los monumentos funeralios. aportación a la historia de la especie. Ochoa C (ed) La fáblica de Anejos de AEspA XLIV, Madrid tabacos de G ijón: arqueología e histOIi a de un espacio milenári o, Raynal JP, Sbihi-Alaoui FZ, Mohib A, E I Graoui M, Lefévre D, Texier JP, Ayuntamiento de Gijon, Gijon, pp 246- 25 1 Geraads D, Hublin 11, Smi th T, Tafforeau P, Zouak M , Grün R, Loget N, Jean Van Den Driessche N (2006) On the origin ofthe strait of Rhodes EJ, Eggins S, Daujeard C, Fern andes P, Gall otti R, Gibraltar. Sediment Geol 188:341- 356. https:lldoi .org/IO. I O1 6/j . Hossini S, Queftlec A (2010) Hominid cave at Thomas quany I sedgeo.2006.03.0 12 (Casabl,Ul ca, Morocco): recent fi ndings and their eontext. Quat Int Leitch V (201 3) Reconstructi ng histOIy through pottely: the contribution 223:369- 382. https:l/doi.org/ 10. 101 6/j .quaint.2010.03 .011 of Roman N African cookwares. J Roman Archaeology 26:28 1- Reimer PJ, Bard E, Bayliss A, Beck JW, B lackweUP G, Bronk. Ramsey C, 306. https:l/doi.org/ IO. I O1 7/S I 047759413000 160 Buck CE, Cheng H, Edward s RL, Friedrich M , Grootes PM, Masseti M (2 009) The mongoose of the cave ofNelj a, southern Spain, is Guilderson TP, Hafl idason H, Hajdas I, Hatté C, Heaton TJ, Hogg not the oldest Egyptian mongoose ofEuro pe. Archaeofauna 18:65- AG, Hughen KA, Kaiser KF, Kromer B, Manning SW, N iu M, 68 Reimer RW, Richards DA, Scott EM, Southon JR, Tumey CSM, Masseti M, Albarella U, De Grossi Mazzorin J (2010) The crested por­ van der Plicht J (201 3) IntCa1l3 and MARTNE I3 radiocarbon age cupine, Hystrix cristata L., 1758, in Italy. Anthropozoologica 45(2): calibration CUlves 0- 50000 years calBP. Radiocarb on, 55(4): 1869- 27-42 1887. do i: https:l/doi.org/I0.2458/azujs_l"C.55 .16947 Modolo L, Salzburger W, Mmtin RD (2 005) Phylogeography ofbarbmy Riquelme-Cantal JA, Simón-Vallejo MO, Palmqvist P, COItés-Sánchez M macaques (Macaca sy /vanlls) and the origin ofthe Gibra ltar colony. (2008) The oldest mongoose of Euro pe. J Archaeol Sei 35(9):247 1- Proc atI Acad Sci U S A 102(20):7392- 7397 2473. https:l/doi.orglI 0. 1O 1 6/j .jas.2008.03 .0 15 MOI·ales I , Pickford M (2008) Creodonts and call1ivores from the middle Rohling EI, Foster GL, Grant KM, Marino G, Roberts AP, Tamisiea ME, M iocene M uruyur fo rmation at Kipsaraman and Cheparawa, Will iams F (201 4) Sea-Ievel and deep-sea-temperature variability Baringo DistTict, Kenya. Comptes Rendus Palevol 7(8):487-497. over the past 5.3 million years. Nanlre 508(7497):477-482. https:// https://doi.org/I0.1016/j.crpv.2008.09.011 doi.org/ l 0.1 038/nature 13230 Morales-Mun iz A (1994) Earliest genets in Europe. Nature 370(6490): Roll mld N (20 13) The Earl y Pleistocene human dispersaIs in the Circum­ 5 12- 51 3. https:l/doi.org/I0.103 8/370512bO MeditelTanean Basin and ini tial peopling ofEurope: single or mul­ Muirhead JR, Gray DK, Kell y DW, Ellis SM, Heath DD, MacTsaac HJ tiple pathways? Q uat Int 3 16:59- 72 . https://doi.org/ 10. 101 6/j. (2008) Identi fY ing the source of species in vasions: sampling inten­ quaint.2013.06.028 sity vs. genetic diversity. MoI Ecol 17(4):1020- 1035. https:l/doi. Smltonja M, Pérez-González A (20 10) M id-Pleistoeene Acheulean indus­ orgll 0. llllIj.1365-294X.2008.03669.x triaI complex in the Iberian Peninsul a. Qual Int 223: 154- 16 1. Muttom G, Scardia G, Kent DV (201 8) Earl y hominins in Europe: the https:l/doi.org/lO.1 O1 6/j.quaint.20 I 0.02 .0 I O Galerian migration hypothesis. Quat Sci Rev 180: 1- 29. https://doi. Santonja M, Pérez-González A, Panem J, Rubio-Jara S, Méndez-Quintas orgll 0.10 16/j.quascirev.2017.1 0.031 E (201 6) The coexistence of Acheul ean and ancient middle

~ Springer Sei Nat (20 18) 105:63 Page 13 of 13 63

Palaeolithic techno-complexes in the middle Pleistocene of the of wi ld boar (Sus sem/à). PLoS One 12(7):eO 18 1929. https://doi. lberian Peninsula. Quat lnt 411 :367- 377. https://doi.org/ 10. 101 6/j. orgll 0.1371/ jOlunal.pone.O1 8 1929 quaint.2015.04.056 Straus LG (200 1) Africa and lberia in the Pleistocene. Quat lnt 75:9 1- Scheidel W (20 14) The shape of the Roman world: modelling imperial 102. https://doi .orgIIO. 10 16/S 1040-6 182(00)0008 1-1 connectivity. ] Roman Archaeology 27:7- 32. https://doi.orgIIO. Van der Made 1 (20 II) Biogeography and climatic change as a context to 101 7/S I 0477594 14001 147 human dispersaI out of Africa and within Eurasia. Quat Sci Rev Shackleton lC, Van AndeI TH, Runnels CN ( 1984) Coastal paleogeog­ 30( 11- 12): 1353- 1367. https://doi.orglIO.1O 1 6/j.quascirev.20 I 0.02. raphy ofthe central and westem Mediterranean dUling the last 125, 028 000 years and its archaeological implications. 1 Field Archaeol Vázquez de la Cueva A (1985) Sigillata afi"icana en Augusta Emerita. II (3):307- 314. https://doi.orgIl0. 11 79/009346984791535476 Monografias emeritenses, Mérida Sharon G (20 11 ) Flakes crossing the straits? Enlame flak es and northem Walker Ml, López-Mmtínez M, CmTión-GarcÍa lS, RodIÍguez-Estrella T, Africa- Iberia contact during the Acheu lean. Afr Archaeol Rev DeI-Toro MS , Schwenninger lL, López-liménez A, Ortega­ 28(2): 125- 140. https://doi.orgIl0.1007/s I0437-011 -9087-3 Rodrigáiiez 1, Haber-Ulimte M, Polo-Camacho l-L, GarcÍa-TolTes Sharon G, Barsky D (20 16) The emergence ofthe Acheulian in Europe­ 1, Campill o-Bo M, Avi lés-Femández A, Zack W (20 13) Cueva A look from lhe ease Quat lnt 411 :25- 33. ht1ps://doi.orglI 0.1016 /j . Negra deI Estrecho deI Río Quípar (Murcia, Spain): a late Early quaint.20 15. 11 .1 08 Pleistocene hominin site with a n "Ache ul o-Levalloiso­ Sil va AM, Boaventura R, Pimenta 1, Delry C, Cardoso lL (20 14) Mousteroid" Palaeolithic assemblage. Qual lnt 294: 135- 159. Perscmtando Espólios antigos: a gmta de Pedra Furada I (Vila https://doi.org/ IO. I O1 6/j.quaint.20 12.04.038 Franca de Xira). Estudos Arqueológicos de Oeiras 2 1: 159- 182 Werdelin L, Dehghani R (201 1) Camivora. ln: HalTison T (ed) Paleonlology Soares 1 (2013) Sal e conchas na Pré-Hislólia pOItuguesa O povoado da and geology of Laetoli: human evolution in context. Springer, Ponta da Passadeira (estuário do Tejo). Setúbal Arqueológica-Pré• Dordrecht, pp 189- 232. https://doi.orglI0.1007/978-90-48 1-9962-4_8 histólia das Zonas Húmidas Paisagens de Sal 14: 17 1- 196 Wood R (2015) From revolution to convention: the past, present and Soria-Boix C, Donat-TOITes MP, Urios V (20 17) Contacts in the last 90, future of radiocarbon dating. 1 Archaeol Sci 56:61 - 72. https://doi. 000 years over lhe sU'a it of Gibraltar evidenced by genetic analysis org/ I 0. 10 16/j .jas.2015.02.0 19

IQ Springer