Direct dating confirms the presence of otter and badger in early Holocene Ireland

ABSTRACT The origin of Ireland’s fauna is an unresolved issue in the history of the island. Although researchers once considered Ireland’s mammals to have spread to Ireland during the early postglacial from 11,700 cal BP, research has increasingly suggested humans translocated many of these species to Ireland. However, due to sparse evidence, the origin of Ireland’s purported native mustelid species remains unanswered. In this study, we shed light on the history of otter and badger in Ireland by directly dating suspected early examples of these species from Derragh, Co. Longford, a late Mesolithic campsite. The results demonstrate that both of these species have been present in Ireland for at least 7000–8000 years. It is not clear if they represent self-colonisers or Mesolithic anthropogenic translocations. Although these finds do not rule out local extinction and later reintroduction, they show that we can no longer assume they are solely late Holocene introductions. GRAPHICAL ABSTRACT


Introduction
The limited number of animal taxa that occur in Ireland has been a source of comment for hundreds of years (Cróinín 2016;Wright 2000;Mitchell and Ryan 2001;Montgomery et al. 2014;van Wijngaarden-Bakker 1989).Differences between terrestrial mammal taxa in Ireland and mainland Europe have been believed to be a result of a peculiar postglacial recolonisation during the deglaciation of northwest Europe that cut off Ireland from mainland Europe far earlier than Britain was cut off from mainland Europe.
Past researchers have speculated that long disappeared late-surviving landbridges may have once existed (Devoy 1995;Yalden 1981).These connections, possibly composed of gravel or ice connecting Ireland to Scotland or Wales, would have allowed fauna recolonisation during the postglacial period (Yalden 1981).Even taking into account the challenges of recolonisation via shifting low-lying landbridges that might restrict some species more than others, this landbridge model fails to explain the unusual composition of fauna found in early Holocene Ireland (Carden et al. 2012;Woodman, McCarthy, and Monaghan 1997).
Oceanographers now believe that post-Last Glacial Maximum sea-level rise cut off Ireland from the island of Britain and continental Europe before 19 kya, long before deglaciation (Corbet 1961;Mitchell 1963;Montgomery et al. 2014).Advances in benthic imaging show that the last landbridge or icebridge connecting the island of Ireland to Scotland and the European mainland were flooded sometime before 15 kya when cold climatic conditions on the island still only supported subarctic mammal fauna (Clark et al. 2012;Edwards and Brooks 2008).This history of biogeographic isolation contrasts sharply with much of northwest Europe which remained accessible for mammalian recolonisation for much longer, for example, the island of Britain remained connected to Continental Europe until to 8 kya or later (Weninger et al. 2008).Ireland's early separation left its mark on the distinct history of human and fauna occupations.The documented human history of the island was shorter than elsewhere in Europe and entirely lacks a Lower and a Middle Palaeolithic.While an Upper Palaeolithic human presence predating the Mesolithic is confined to rare examples of worked faunal bone (Dowd and Carden 2016;Woodman 2015).Rising seas resulted in a notably lower number of mammalian recolonisation events than elsewhere in northwest Europe.Current scholarship implies a "prey gap", an ecosystem that was disproportionately rich in carnivores (wolf (Canis lupus), bear (Ursus arctos), lynx (Lynx lynx), wild cat (Felis silvestris), stoat (Mustela ermine)) with few herbivores (wild pig (Sus scrofa), hare (Lepus timidus), and wood mouse (Apodemus sylvaticus)) (Kitchener 2012;Sleeman 2008).
The lack of any landbridges or icebridges in a climate window suitable for recolonisation of temperate-type non-flying terrestrial mammals means that the origin of Ireland's fauna has been an unsolved question for biogeography and archaeology studies (Edwards and Brooks 2008).A number of these species such as the mountain hare, Eurasian stoat, otter (Lutra lutra), and badger (Meles meles), exhibit morphological and behavioural traits unique to Ireland, suggesting a longer split from European populations (Dadd 1970;Lynch 1996;Lynch and O'Sullivan 1993;O'Sullivan, Sleeman, and Murphy 1992;Sleeman and Davenport 2016).
There is evidence to conclude that several fauna is either cold-tolerant pre-Holocene survivals, i.e. stoat, mountain hare and the now extirpated brown bear (Ursus arctos), or were early Holocene recolonisers from cryptic refugia.However, there is insufficient archaeozoological evidence of many species including otter, badger, red fox (Vulpes vulpes), pine marten (Martes martes) and red squirrel (Sciurus vulgaris).This has resulted in research placing an increasing Table 1.Earliest evidence of Irish mammal species, excluding bats as they may have used airborne dispersals.Three categories of fossil evidence are listed including the least strong e.g.culturally dating, which is subject to erroneous associations due to layer remixing, and strongest (directly dated), in addition genetic evidence is presented.Mascheretti et al. (2003) emphasis that many of these species are a more recent part of the Irish landscape than once thought (Table 1) (Allen et al. 2020;Frantz et al. 2014;McDevitt et al. 2011;O'Meara et al. 2012;Woodman 2012;Yalden 1999) .Fossil taxa that are dated by their association with culturally and radiocarbon dated deposits often represent intrusive finds from more recent periods.This is particularly likely given that many of these taxa including otter and badger, either are burrowing mammals or reuse existing burrows, both of which frequently disturb and mix archaeological layers.Researchers have, using the fossil and the genetic record, speculated that many of the taxa once assumed to be native, may be early to mid-Holocene introductions, brought to Ireland by boat for fur or other purposes (Lynch 1996;Lynch and O'Sullivan 1993;Mallory 2013;Montgomery et al. 2014;Woodman 2008).In some cases, such as the badger, researchers have suggested an introduction as recent as the medieval period (Monaghan 2017) and while genetic data has been interpreted as suggesting both natural colonisation (O'Meara et al. 2012) and colonisation through a series of waves with human aid via Britain (Allen et al. 2020;Guerrero et al. 2018).
While the otter has been suggested to self-recolonise during the Bronze Age (Searle 2008).These interpretations challenge their status as native mammals and may impact how ecologists design conservation policy around these species.
To elucidate the origin and phylogeographic history of Irish mustelids, we applied direct 14 C dating, and dietary stable isotope analysis on early mustelids remains that could be possibly intrusive to the Derragh archaeological site.These include an unmodified cranium identified as otter morphologically and an unmodified badger tibia identified morphological and with ZooMS from early Holocene deposits at the lakeside site of Derragh in Co. Longford (Fredengren 2007;McCarthy and McCormick 2018;Perri, Power, Stuijts, et al. 2018;Spindler 2018).We then performed species accumulation analysis, a common methodology in ecology to measure representativeness of sampling (Gotelli and Colwell 2001), on the number of taxa recovered from all mammal assemblages to document, if sufficient zooarchaeological remains have been found to estimate Mesolithic faunal diversity in Ireland.

Site and materials: site background, stratigraphy, and dating
The Derragh site (Irish: Doire Each; Co. Longford; 53.772, −7.403) is a former lake island, located at Derragh lough in the Irish Midlands, about 66 km from present-day sea shorelines.It is currently on a lakeshore, lying at the junction of Lough Kinale, Derragh Lough and the outlet of the River Inny, a tributary of the River Shannon.The archaeological potential at the site was detected after exposed Bann flakes were discovered in the 1970s (O'Sullivan 1998).The full extent of the site was exposed following lake drainage programmes.The Discovery Programme surveyed the site in 2002 and excavated over several weeks in the summers of 2003-2005(Fredengren 2002;;2007;2009;Fredengren, Kilfeather, and Stuijts 2010).Excavation of the site identified a stake structure, a hollow, three hearths and ash and hazelnut spreads dating to 7424-7020 cal BP.Worked oak planks overlaid this layer, which was sunk under more hazelnut shell spreads dating to 6890-6650 cal BP.Above these deposits lay a platform-like layer of stones with abundant archaeological evidence, including lithics, animal bones, including modified bones of wild pig and bear, dating to 6950-5970 cal BP (McCarthy and McCormick 2018;Perri, Power, Stuijts, et al. 2018).Lithic specialists argue that the weathering of the tools in this layer indicates secondary deposition (Sternke and Woodman 2006).The earliest date from primary deposits is a piece of oak charcoal underneath this layer dated 7620-7670 cal BP (McCarthy and McCormick 2018;Perri, Power, Stuijts, et al. 2018).Aside from some probable lithic evidence of Early Mesolithic occupation, the overwhelming amount of artefactual evidence suggests a series of ephemeral Late Mesolithic to Early Neolithic occupations (Fredengren 2002;2007;2009;Perri, Power, Stuijts, et al. 2018;Perri, Power, and Heinrich 2018;Power 2018).

Methods
The bone samples (B493 and B563; Supplementary figure 2, 3) were pretreated at the Department of Human Evolution at the Max Planck Institute for Evolutionary Anthropology (MPI-EVA), Leipzig, Germany, using conventional methods (Talamo and Richards 2011).The outer surface of the bone samples was first cleaned by a shot blaster, and then 500 mg of the whole bone was taken.The samples were then decalcified in 0.5M HCl at room temperature until CO 2 effervescence ceased.Subsequently, 0.1M NaOH was added for 30 min to remove humics.The NaOH step was followed by a final 0.5M HCl step for 15 min.The resulting solid was gelatinised (Longin 1971) at pH3 in a heater block at 75°C for 20 h.The gelatine was then filtered in an Eeze-Filter™ (Elkay Laboratory Products (UK) Ltd.) to remove small (>80 μm) particles.The gelatine was then ultrafiltered (Brown et al. 1988) with Sartorius "VivaspinTurbo" 30 KDa ultrafilters.Prior to use, the filter was cleaned to remove carbon containing humectants (Brock, Ramsey, and Higham 2007).The samples are lyophilised for 48 h.All dates were corrected for a residual preparation background estimated from 14 C free bone samples.These bones were kindly provided by the Mannheim laboratory and pretreated in the same way as the archaeological samples (Korlević, Talamo, and Meyer 2018).
The preservation of the collagen was ensured by only evaluating samples with a yield of collagen not less than 1%, C:N ratios (between 2.9 and 3.6), together with isotopic values were (van Klinken 1999).The isotopic values were obtained as a single run at MPI-EVA (Lab Code R-EVA) on a ThermoFinnigan Delta V Advantage isotope ratio mass spectrometer (IRMS) coupled to a Flash 2000 EA.Stable carbon isotope ratios are expressed relative to VPDB (Vienna PeeDee Belemnite) and stable nitrogen isotope ratios were measured relative to AIR [atmospheric N2 with an analytical error of 0.2‰ (1σ) for δ 13 C and δ 15 N].After passing the quality criteria between approximately 5-6 mg of collagen was weighed into pre-cleaned tin cups and sent to the Klaus-Tschira-AMS centre, Mannheim, Germany (lab code: MAMS).Samples were then combusted in an Elemental Analyser and CO 2 converted catalytically to graphite and dated using the MICADAS-AMS (Kromer et al. 2013).
Lastly, we conducted sample-based species accumulation analysis (collectors curve) with a logarithmic function on the number of mammalian species detected in Mesolithic zooarchaeological assemblages, including land and marine mammals to ascertain if current diversity estimates are reliable.To do this we sourced the number of archaeological sites with in situ and reworked Mesolithic deposits from published sources including the database of Irish Excavation Reports (Excavations.ie.).Reports with deposits described as "possibly Mesolithic" were omitted but sites described as "Mesolithic" or "probably Mesolithic" were included (Supplementary table 1).

Radiocarbon dating
The results place the two specimens at 95.4% probability (otter = 7688-7588 cal BP, badger = 7158-6951 cal BP) firmly in the Late Mesolithic period, corresponding to cultural and 14 C dates from this layer (Table 2).The otter is significantly older than reported dates from hazel brushwood from the same context 65 which overlays context 93 (Ua-32383 = 7470-7310 cal BP).It is also considerably older than the earliest date of the sequence, oak charcoal under the stone layer (GrN-26351 = 6480 ± 40 BP) (Fredengren n.d.).Although much older giant deer fragments occur at the site and are found in a layer above context 65, their human modification despite being extinct in the Mesolithic and direct dating show these represent secondary deposition (UBA-26786 10199 ± 49 BP) (Fredengren n.d.).The incongruously early date of the otter is probably linked to the "Old Carbon Effect".The freshwater Old Carbon Effect is poorly understood and its impact may be highly variable.Without more environmental information, it cannot be corrected for in this assemblage in the scope of this project; however, most research suggests its effect may only distort a date a few centuries and an age shift of 1200 uncal years as observed is improbable (Hart et al. 2019

Carbon and nitrogen stable isotope ratio dietary reconstruction
The δ 13 C and δ 15 N results are presented in Table 3 and plotted in Figure 3. Their %C, %N and C:N elemental ratio values show sufficient collagen survival to infer dietary values (Ambrose and Norr 1993;van Klinken 1999).The otter's δ 13 C and δ 15 N values, a carnivore, are −23.1‰and 10.5‰, are consistent with a diet that is contrasting with other studied taxa and shows it obtained its protein from a different source, likely from aquatic animals.The badger's δ 13 C and δ 15 N values are −23.57‰and 11.36‰.Badgers are omnivorous and eat a high diversity of food items.The badger's diet is trophically similar to the Rockmarshal Mesolithic forager.It is likely to have had an omnivorous diet comprised of terrestrial plant and animal foods.

Species accumulation analysis
Collation of published Mesolithic site data identified 115 sites on the island of Ireland with evidence of Mesolithic occupation.30 sites with mammal assemblages were identified in which nine species were present.The resulting curve does not plateau (Figure 4).This indicates that current sample sizes of mammals on Mesolithic sites are not too limited to calculate Mesolithic species diversity.This indicates that mammal diversity is currently a function of sampling effort.

Discussion
The radiocarbon dates show that badger and otter were present in Ireland from at least the Later Mesolithic, and that otter may have been present from the end of the Early Mesolithic.The remains of otters, badgers and other mustelids are rare on Irish archaeological sites.Only the remains of stoat has been so far identified in Mesolithic deposits and directly dated, but it is not clear if they represent an anthropogenic or a natural deposit (Woodman, McCarthy, and Monaghan 1997).However, if Mesolithic foragers only valued this species for fur, they may have processed with out-of-camp butchery, resulting in less skeletal remains deposited on archaeological sites.However, elsewhere in the Mesolithic world, mustelid remains are less rare, for example, at Jutland where pine marten and otters were a large proportion of the faunal assemblage (Andersen 1973).Across Mesolithic Europe, pine marten with cut-mark evidence of skinning is frequent (Llorente-Rodríguez et al. 2016).Unlike other regions of Europe, Mesolithic faunal assemblages in Ireland have an extremely low number of identified specimens, and this contributes to the difficulty detecting these species in Ireland.The species accumulation curve shows that faunal assemblages have low species diversity and that although wild pigs will not be missed, our knowledge of rarer Mesolithic prey, non-staples is heavily influenced by chance finds (Figure 4).This hints that new fieldwork that identifies early Holocene zooarchaeological assemblages could yield so far unrecovered species.
The unmodified nature of the remains means that we cannot identify if Mesolithic people hunted otter and badger at Derragh; while the early dates could mean that these remains were earlier deposits which were being used as a utilitarian source of materials or for ritualistic purposes.Even more parsimonious is the possibility that they simply are part of reworked deposits containing natural remains.It is already clear that at Derragh and other contemporary Irish Mesolithic sites, people retrieved and modified skeletal remains of the Pleistocene species giant deer (Megaloceros giganteus), which was already extinct for several thousand years (Warren 2017;Woodman 2012).Occupants of Mesolithic layers chopped the back of the skull for unknown reasons (McCarthy and McCormick 2018).Given that the otter skull may have been hundreds of years old when deposited, it is possible that the occupants of the site curated old otter and giant deer skulls for symbolic purposes, such as totems or it could simply represent an otter that died naturally.Although badgers are unlikely to occupy the immediate lakeshore environment due to their dislike of waterlogged environments, the otter would have been at home in the immediate surrounds which likely offered salmonids and eel (Perri, Power, and Heinrich 2018).Unlike otters, badgers avoid digging setts in waterlogged environments and the Derragh badger remain seems likely to have been brought to the site.Although Mesolithic transport and curation of a wide variety of skeletal elements from small fauna are well known, motives are often unclear (Overton 2014;2016).The dietary analysis provides lessons on the dietary ecology of badger and otter.The otters depleted δ 13 C bone collagen stable isotope values show the otter probably consumed a freshwater diet, which is clearly distinct from that of extant human Mesolithic individuals.The values are similar to other studied Eurasian otters and indicate anadromous fish were not an important part of its diet (Drucker and Bocherens 2004).Its diet may have included non-anadromous salmonids, eel, stickleback, freshwater mussels and invertebrates, but it cannot be clarified (Ottino and Giller 2004).Badger δ 13 C and δ 15 N values show an omnivorous diet similar to human values identified with inland Mesolithic people at Killuragh Cave, Co. Limerick and Loughan Island and the coastal site of Rockmarshall, Co. Louth, in contrast to the marine Mesolithic diets evident from enriched δ 13 C at Ferriter's Cove, Co. Kerry.Past studies of stable isotopes of badgers have provided evidence of similar diets, as well as contrasting low trophic level diets too (Robertson et al. 2015;Villalba-Mouco et al. 2018).Badgers ecosystem engineers who influence nutrient cycling, and seed dispersal, particularly of large-seeded fruit and this may have implications for understanding other species on this island (Kurek et al. 2022).
The Derragh otter and badger bring the number of species present since the Mesolithic period to at least nine, even though genetic evidence based on DNA and morphology has been interpreted as suggesting that several of these were later arrivals.These lines of evidence may not be contradictory.Present-day populations of badger and otter may represent more recent colonisations that replaced, or admixed with initial founder populations or colonised empty landscapes following local extinction.Thus, extant populations of otters and badgers are not necessarily descended from the populations represented at Derragh.Cycles of colonisation and extinction have been suggested for Irish species such as the red squirrel (Fairley 1975).As a carnivore and omnivore species respectively, the addition of these two Mustelid species to lists of Mesolithic fauna do not help to explain the "prey gap", the seemingly disproportionately balance of predators (eight) to prey (four).In many cases, evidence of later arrivals is scant, and the apparent presence of a "prey gap" in the early Holocene may be due to insufficient archaeozoological assemblages (Kitchener 2012;Sleeman 2008).Ecologists have predicted that further prey species will be detected (Sleeman 2008).
The presence of otter and badger in Early Holocene Ireland has implications for biogeographical models that compare similarities of Irish fauna with other European faunas (Montgomery et al. 2014).It is possible that otters were translocated to Ireland, perhaps as cubs to provide fur for Mesolithic peoples but this is entirely conjectural.It has been observed that Irish otters often show white pelage spots, which is a sign of domestication (Sleeman, pers. comm.)(Scharff 1906b).Researchers have already suggested that human manipulation of wild species in Ireland may have begun even prior to the adoption of farming and has been suggested in the case of wild pig, land snails and potentially all species aside from stoat and hare (Grindon and Davison 2013;Warren 2013;Woodman and McCarthy 2003).Human management may have extended beyond simple translocation, and it has been suggested that Mesolithic hunters conserved wild pigs through avoidance of hunting mature females (McCormick 2004).However, self-colonisation of mammals into Ireland must be considered as a serious possibility.Red fox, often thought to have been an ancient human-mediated species, has recently been suggested as an early post-glacial colonisation based on analysis of their genetic structure (McDevitt et al. 2022).Closely related species of otters have been observed to forage in marine waters up to 13 km offshore, while Scotland today only lies about twice this distance from parts of northeast Ireland, at the beginning of the Holocene it was several kilometres less (Blundell, Ben-David, and Bowyer 2002;Finnegan and Néill 2010;Woodman 2015).Recolonisation from cryptic refugia may have also occurred.Otters probably were readily able to self-colonise through dispersing into Ireland from the narrow sea channel between Scotland and north eastern Ireland.Badger dispersal over sea barriers is harder to understand.Red fox, often thought to have been an ancient humanmediated species, has recently been suggested as an early post-glacial coloniser about 16,000 years ago from Britain, based on analysis of their genetic structure (McDevitt et al. 2022).While red foxes may have self-colonised through routes open to badgers, they would have intensely cold open environments.Red foxes are more tolerant of cold environments than badgers, and would be a better candidate to disperse at this early date.Badger swimming ability is poorly documented but is believed to occur and North American badgers have been observed swimming 800 m from lakeshores (Duebbert 1967;Kirkwood 2009;Sleeman, Cussen, and O'Leary 2002), while Irish badgers are known to swim and cross beaches (Sleeman, O'Leary, and Cussen 1999).It is unknown if Eurasian badgers could swim further distances, and if this possible in saltwater conditions.For badgers, human translocations may still be argued to represent a highly plausible origin.Nor can we rule out self-recolonisation from cryptic refugia.Released badgers would have represented a source of meat and fur to Mesolithic foragers who aside from wild pigs were largely bereft of terrestrial meat-bearing mammals.Although historical evidence would suggest that badger was not a major source of fur and meat in traditional Irish societies, there is evidence of their occasional use, i.e. reference to saill bruicc a Bérre "salted badger meat" in a twelth-century poem (Kelly 1997).An early seventeenth century text states that "its flesh is not unpleasant to eat nor its pelt to be despised.It is reported that it is helpful against the spell of the evil eye.Because of this, tongs of hair made from the skin are added to the bridles of baggage animals" (O'Sullivan 2009).Another Irish source (Rutty 1772) notes that "the flesh, when roasted, is good food, like pig's flesh, and makes a good ham; and the skin is tanned for breeches, waistcoats, etc." Consumption of badger meat has been cited as motivation for translocation in the Mediterranean Neolithic world.In the Mediterranean it is believed badgers were introduced onto Crete by Neolithic peoples (Groves 1989;Vigne 1999) and cut marked remains testify to the regular use of this species.

Conclusions
The directly dated remains transform our knowledge of the presence of badger and otter in the Holocene.The results show that these species were present in Ireland since at least 7000 BP, and that Mesolithic faunal diversity was higher than thought.It is plausible that otters self-colonised Ireland, but it is less clear if badgers may have self-colonised.Due to the difficulties in self-colonisation, badgers may have been sufficiently valued to have been translocated to the island.
While the addition of this carnivore and omnivore, do not solve the "prey gap", they indicate a more complex ecology that was more similar to other regions of Europe.While isotopically the otter displays a distinct and likely freshwater diet, the badger remains indicate a more mixed diet including plants and animal foods, tropically comparable to Irish Mesolithic foragers.Mesolithic people could have hunted these species, and brought their remains to the site.The wet nature of the environment suggests that at least badger remains was introduced to the Derragh site by people, but due to the lack of cutmarks this cannot be clarified.Prof. Sahra Talamo is a professor in the Department of Chemistry CHIM in the University of Bologna where she runs a radiocarbon dating laboratory.She is interested in refining chronologies of human expansions in Eurasian as well as the population history of archaics.

Figure 1 .
Figure 1.Location of sites with extant evidence of mammals mentioned in the text.

Figure 2 .
Figure 2. Location of Derragh, Co. Longford.Inset map on the top left shows the location of Derragh on the island of Ireland.

Figure 3 .
Figure 3. Carbon (δ 13 C) and nitrogen (δ 15 N) isotope ratio values from otter and badger remains with other fauna and humans values from the Irish Mesolithic.

Figure 4 .
Figure 4.A species accumulation curve with a logarithmic function of the number of identified mammals detected on Mesolithic sites (12 species) indicates there is a likelihood of undetected early Holocene species.Includes seal and dolphin.This analysis does not take into account of multiple examples of a species on a site.
Dr. Robert C. Power is a postdoctoral archaeologist at the School of Archaeology, University College Dublin.His research investigates human diet and adaptation in Eurasia and Africa.Dr. Ingelise Stuijts is an independent environmental archaeologist who is affiliated with the Discovery Programme of Ireland.She specialises in reconstructing environments and human behaviours with pollen, wood and charcoal in prehistoric and historic Ireland as well as in other regions.Dr. Finbar McCormick is a retired senior lecturer in the School of Natural & Built Environment in Queens University Belfast.His principle area of research is the settlement and economy in Early Medieval Ireland but he also has a long-standing interest in the mammalian history of Ireland.

Table 2 .
Fauna remains analysed by the study.