5

Mammal Bone Studies from Prehistoric Irish Sites

Finbar McCormick

Abstract

Our knowledge of the utilisation of mammals in prehistoric Ireland has greatly expanded during the past four decades, especially since the publication of van Wijngaarden-Bakker’s pioneering work on the Beaker material from Newgrange in 1974. This essay summarises the knowledge that has emerged since then spanning the period between the Mesolithic and Iron Age. It also outlines the major gaps that exist in the record, the most obvious being for the Neolithic and Late Iron Age.

Introduction

Zooarchaeological studies in Ireland date back to the middle of the nineteenth century but it is only during the last 30 years that standardised methods of quantification, metrical analysis and the recording of ageing data have allowed a more rigorous discipline to emerge. Ireland, however, can lay claim to much pioneering work during the early days of faunal analysis. Molyneaux’s (1697) early deliberation on the remains of the Giant Irish Deer is one of the earliest works to deal with megafaunal extinctions, recognising that extinction must have occurred at least on a regional scale. William Wilde’s (1840; 1857–61) articles discussing the animal remains from Lagore Crannog, Co. Meath, and other archaeological sites must be amongst the earliest to deal in a detailed way with zooarchaeological material. More recently, Margaret Jope’s work on faunal material from excavated sites in Counties Antrim and Down during the 1950s are amongst the first zooarchaeological reports to tabulate the faunal remains from excavations and develop a systematic method for estimating the minimum numbers of individuals (e.g. Jope 1954; 1955). The work of A. W. Stelfox and Geraldine Roche, both based in the Natural History division of the National Museum, Dublin, also provides much useful information on faunal material from excavations conducted in Ireland during the middle of the twentieth century. They were especially involved with the material retrieved from the crannog excavations undertaken in Ireland by the Harvard investigations that produced large quantities of material (Roche and Stelfox 1936; Stelfox 1942). Their interests, however, lay more in the zoological rather than the archaeological fields and their reports are of limited value today. This is more than compensated for, however, by the fact that they retained much important material including skulls, complete long bones, and generally the bones of wild species. These are presently curated and available for study in the Natural History Museum collections in Dublin.

Methodology

Because of the absence of standardised methods of quantification early reports are of limited value when it comes to inter-site comparisons. Furthermore, many excavation reports pay insufficient attention to site stratigraphy and this, often combined with poor dating, lessens the value of many of the early faunal reports. If any work could be identified to mark the beginning of zooarchaeology in Ireland as a systematic discipline, it must be Louise van Wijngaarden-Bakker’s first report on the material from the Beaker levels at Newgrange, Co. Meath, published in 1974. Since then, a series of reports using standardised methods of recording have greatly contributed to our understanding of the exploitation of animal resources in Ireland’s past.

An acceptable faunal report needs to contain some basic information, much of it in tabular form. It should contain tables outlining the main NISP (number of species parts) and MNI (minimum number of individuals) values for the main contexts/phases from the site. The basic ageing data for both tooth eruption and wear, and epiphyseal fusion data must be presented in such a way that it can be used for comparison with other sites. At the very least, a summary of the metrical information should be presented and complete bone dimensions, especially those of the less frequently encountered species, should be recorded individually. A report should also record any palaeopathological lesions and butchery marks encountered.

Animal Exploitation in Prehistoric Ireland

The present case study evaluates the nature of animal exploitation in prehistoric Ireland based primarily on the study of faunal material during the last 30 years. Prior to 1970, relatively little was known about the subject and some of the assumptions made were erroneous. The Mesolithic economy, for instance, was believed to have been based primarily on coastal shellfish exploitation, supplemented by fishing, fowling and the occasional exploitation of mammals, especially deer. Major aspects of this model are now known to be incorrect. Large oyster middens, for instance, are now known to be a Neolithic phenomenon, while the wild pig was the main component of the Mesolithic meat diet, the red deer being absent from the country. With the exception of a small quantity of Late Mesolithic/Early Neolithic material from Lough Gur, Co. Limerick (O’Shaughnessy forthcoming), there is virtually no faunal data from any early post-Mesolithic settlement sites. While the Neolithic is essentially still a mystery as far as animal exploitation is concerned, we now have an emerging understanding of the livestock economies and meat diets of the Bronze and Iron Ages.

 

Species Availability

The range of mammals available for exploitation on the arrival of humans was extremely limited. Table 5.1 compares the mammals present before, and after, the Younger Dryas

	Species
Pre-Younger Dryas approx. 12,000-10,000 BP	Giant Irish Deer
	Red Deer
	Arctic Lemming
	Reindeer
	Bear
	Stoat
	Hare
Post-Younger Dryas approx. 9000-7000 BP	Wild pig
	Bear
	Wild Cat
	Hare
	Canis sp.
	Wood mouse
	Lynx

Table 5.1: Early Holocene species recorded in Ireland (after van Wijngaarden-Bakker 1989; Woodman et al. 1997).

period. A range of animals disappeared either during the Younger Dryas, when Ireland reverted to sub-arctic conditions, or in the warm period that immediately preceded it. Only the bear and the hare were definitely present both before and in the few millennia post-dating the Younger Dryas. The unspecified canid bones from some Early Mesolithic sites are probably wolf so this species can also be added to the list. Were these few mammal species Late Glacial survivors who colonised Ireland via a land bridge before the Younger Dryas and then managed to survive that cold period? Alternatively, were all the extant animals killed off by the sudden cold period and subsequently introduced to Ireland by humans? If humans are capable of introducing large mammals, such as domesticated cattle, it is just as likely that they were capable of importing large and small wild species, especially as young animals. Even bears could have been imported as there is evidence on continental Europe that they were tamed during the Mesolithic (Chaix et al. 1997). The wild pig was the principal species hunted in Mesolithic Ireland and as a forest animal it must have been an Early Holocene introduction rather than a Late Glacial relict. Therefore, at present, there is at least a strong possibility that all of Ireland’s mammals are deliberate or accidental human introductions. The range of freshwater fishes was also much more limited than in Britain or continental Europe due to Ireland’s isolation after the last Glaciation when re-colonisation of fresh-water species was not possible (McCormick 1999).

 

Mesolithic

During the first three quarters of the twentieth century a small series of midden sites had produced faunal remains, but the dating of this material was problematic as there was often an overlap with deposits of the Early Neolithic. The first sites to have produced unequivocal Mesolithic faunal material were at Mount Sandel, Co. Londonderry and Lough Boora, Co. Offaly, both of which date to the Earlier Mesolithic (van Wijngaarden-Bakker 1989). While only calcified bones survived at both sites, they reflect a general mammalian exploitation strategy that was to persist throughout the Mesolithic. Red deer were absent and the exploitation of terrestrial mammals was almost solely dependent on the wild pig (Table 5.2; Fig. 5.1). Whether or not the animal protein diet was dominated by wild pig, however, is unclear. At both these sites relatively large quantities of fish bone were recovered; trout and eel at inland Lough Boora with a more diverse range, dominated by salmon and eel but also including sea bass and flatfish, having been exploited at Mount Sandel. This wider species list is a refection of the site’s location at the mouth of the River Bann. Smaller quantities of bird were also present at both sites but these seem to have played an incidental role in the sites’ economies. Isotopic analysis of human remains from Late Mesolithic sites on continental Europe (Schulting 1998), and more recently in Scotland, has shown that, on coastal sites at least, maritime resources predominated in the diet (Schulting and Richards 2002). Human remains from Late Mesolithic Ferriter’s Cove, Co. Kerry, show an extremely heavy reliance on maritime resources but analysis of human bone from coastal Rockmarshall, Co. Louth, suggests a much less important role in the diet (Woodman et al. 1999, 142–3).

Table 5.2: Mammal species from recently excavated Irish Mesolithic sites (after van Wijngaarden-Bakker 1989, 127; McCarthy 1999, 90; McCormick 2004).

Figure 5.1: Distribution of mammal remains from Irish Mesolithic sites based on the data contained in Table 5.2. The canid was from Mount Sandel while the wild cat was from Lough Boora. NB: the possible canids from Lough Boora were excluded from the graph since their identification was not definite.

There is little evidence for permanent settlement in Mesolithic Ireland, with the mammal bones found being interpreted as seasonal indicators of occupation. For example, the presence of large numbers of unworn pig milk teeth at Lough Boora has suggested summer occupation, an observation supported by the large numbers of immature eel also present (van Wijngaarden-Bakker 1989, 127, 129). Epiphyseal fusion data of wild pig phalanges at Mount Sandel suggested that the animals were about 18 months of age at time of death. Assuming that birth occurred in May, this suggests winter occupation, a conclusion supported by the presence of the red-throated diver, a species that presently occurs in Ireland only between October and April (van Wijngaarden-Bakker 1985, 72). As previously stated, large quantities of salmon were present at Mount Sandel and van Wijngaarden-Bakker suggests that these reflect the main salmon run in the River Bann in July and August. She concluded that there were two distinct periods of occupation at the site (van Wijngaarden-Bakker 1985, 74–5)–winter occupation was centered mainly on hunting wild pigs, while the summer occupation concentrated on the exploitation of salmon.

Pig too predominated on the two Later Mesolithic sites that have recently produced animal bone–Moynagh, Co. Meath (McCormick 2004) and Ferriter’s Cove, Co. Kerry (McCarthy 1999). The assemblage from Moynagh may reflect a degree of specialisation in animal exploitation; despite the site’s lakeside location and optimal waterlogged preservational conditions there was no evidence for the utilisation of fish or bird. Interestingly, fish were also absent from the Late Mesolithic lakeside settlement at Starr Carr, Yorkshire, accompanied by limited evidence of bird exploitation (Frazer and King 1954). It seems that Mesolithic hunters exploited what was easiest to catch and often ignored other resources. Mammalian ageing data at Starr Carr suggests late spring/ summer occupation which is also the best time for exploiting freshwater fish (Frazer and King 1954). Their absence cannot therefore be adequately explained by seasonal unavailability. Zvelebil (1995) has indicated that there is an increased dependence on wild pig in Late Mesolithic Scandinavia and suggests there is also evidence for increased management of pigs at this time. It may well be that the Moynagh assemblage, with its restricted range of animals, is reflecting a similar trend (McCormick 2004).

The Ferriter’s Cove faunal assemblage, along with the human bone isotopic data, indicates an economy highly dependent on marine resources. A much wider range of fish species were present than at Mount Sandel, reflecting its location directly on the coast. Large quantities of shellfish were also present. Most surprisingly, domesticated cattle and sheep were also present in Late Mesolithic contexts, a cattle tooth providing an uncalibrated date of 5510±70 BP (4500–4180 BC cal. BC) (Schulting 1999, 219). This is the earliest dated domesticated bovine from either Britain or Ireland and it may represent an early, and unsuccessful, attempt to introduce agriculture into Ireland from continental Europe.

Table 5.3: Cattle, caprovine and pig bones from Irish Neolithic Sites (McCormick 1985; 1987; 1988; A. Lynch pers comm.). The Poulnabrone values are provisional and based on a preliminary examination.

 

Neolithic

Because of the dearth of Neolithic faunal material, very little is known about the exploitation of animals, domesticates or wild, during this period (Table 5.3). The general absence of ditched enclosures and rubbish pits for Neolithic settlement sites has meant that little in the way of domestic refuse has survived. The faunal remains from a Neolithic house at Tankardstown, Co. Limerick (McCormick 1988, 182–4), are typical of the poor quality of such assemblages. The bones were recovered from the foundation trench, were all calcified and tell us little more than that cattle, pig and caprovines were exploited. Caprovine remains predominated. In contrast, caprovines are absent from small samples recovered from Neolithic contexts at Knowth, Co. Meath (McCormick 1997a). However, such samples are too small to produce reliable results.

Most of the more extensive Neolithic samples of animal bones come from funerary contexts but such material can be extremely problematical. Many Neolithic tombs remained open, or were reused over a large period of time, and it is often difficult to accurately date faunal material or prove its association with burial ritual. Wild animals often used Neolithic chambers and their cairns as lairs. Bones may therefore represent animals that died in and around the tombs or food that had been brought to the site by carnivores. Many of these problems can only be addressed by the expensive process of radiocarbon dating individual animal bones. Furthermore, much of the material has not been retained. When comprehensive radiocarbon dating of Neolithic funerary faunal remains has been undertaken, as was the case for Point of Cott on Orkney (Barber 1997), it was found that animal deposits were often not contemporary with the human deposits in the tomb and were a produce of natural depositional processes.

Faunal remains have been found in many Irish funerary monuments, but in most cases it is impossible to ascertain if they are primary deposits or to assign a date to their deposition (McCormick 1986). It is not possible, for instance, to assign dates to any of the large quantity of faunal material recovered from Ballycarty, Co. Kerry (Connolly 1999). The morphology of the Linkardstown tombs, however, did not generally enable secondary access to the burials and the cairns were often sealed with a thick layer of soil. At Ashleypark, Co. Tipperary, the chamber and cairn were sealed with clay in this way and the radiocarbon dates confirmed a Neolithic date for the animal bones (McCormick 1985). Two types of ritual were noted at the site. The first involved the placing of single bones each of cattle, pig and caprovine alongside the human remains, perhaps token deposits of the main livestock species. The second activity is indicated by a larger sample, almost exclusively of mature cattle, found thrown among the cairn stones that sealed the chamber. This could be interpreted as refuse from a meal, or meals, consumed during the sealing process. The deposition of cattle bone during a sealing process was also noted at the court tomb at Goward, Co. Down (Davis and Evans 1933, 104). No animal bone was associated with the primary tomb deposits on this site but the main chamber had been subsequently partially filled with stones and sealed by a dry stone wall that the excavators regarded as having been of the same phase of activity. Associated with the partial filling of the chamber was the deliberate inclusion of a large number of cattle deposits. The excavators note that ‘there were twenty-three in all sometimes consisting of one bone, sometimes of a number … it was particularly noticed that none of the bone deposits were crushed by superposed stones, but were carefully laid in the cracks among stones’ (Davis and Evans 1933, 104). Unfortunately, no dating evidence exits for this sealing occurrence.

Occasionally, faunal deposits in funerary contexts suggest that the rituals undertaken were similar to those noted outside Ireland. One of the chambers in the dual court tomb at Audleystown, Co. Down, contained cattle teeth and phalanges. If these are of a Neolithic date they may represent ‘head and hoof’ deposits that were common in Britain, Europe and western Asia during early prehistory (Piggott 1962; Ashbee 1970; Robertson-Mackay 1980; Mallory 1981). At Audleystown too, a pig mandible was placed next to that of a human (Collins 1954, plate 5). The placing of pig jaws in chamber entrances has been noted in long barrows in Gloucestershire and Yorkshire in England (McCormick 1986, 38). At Hanging Grimston long barrow several deposits of pig mandibulae were present, one containing at least twenty individuals. Deposits of pig mandibulae are also common in megaliths in Central Europe. In the Lengyel Culture cemetery at Zengovarkony, Hungary, for example, the human skulls of some of the richer inhumation burials appeared to have been replaced with a pig mandible (McCormick 1986, 38).

While molluscs are discussed in detail elsewhere (see Murray, this volume) it is perhaps relevant that their use in funerary ritual contexts is considered here. A large oyster shell was included with one of the primary burials (No. 8a) at Poulawack Cairn, Co. Clare (Hencken 1935, 208–9), while two Venus shells accompanied human bone in the ‘pseudo-megalith’ at Cahirguillamore, Co. Limerick (Hunt 1967, 31). Neither of these are coastal sites, so their inclusion can be accepted as deliberate. More problematic is the presence of approximately 300 limpet and cockle shells, and a whale bone, in the hilltop passage tomb at Loughcrew, Co. Meath, which is nearly 60 km from the nearest coast (Conwell 1873, 51–2). Unfortunately the tomb was much disturbed and included Iron Age and Early Christian artifacts so the date of the marine material cannot now be established. Perhaps the most obscure use of marine shells in a Neolithic burial context is noted at Dalkey Island, Co. Dublin (Liversage 1968, 103–4). On this site some 50 periwinkle shells were found within the brain case of a human burial, dated to ‘2,300 BC, with a standard deviation of 150 years’ (Liversage 1968, 103). The excavator suggests that the shells had been inserted into the skull for ‘superstitious reasons’. The burial was situated within a shell midden but as this comprised mostly limpets, it precludes a natural explanation for the presence of the periwinkles within the cranium.

While faunal remains have provided interesting glimpses into burial ritual in Neolithic Ireland, we remain in some ignorance regarding the livestock economy of the period. Neither do we have much metrical information that could inform us about the type of domesticates present. This is especially unfortunate in the case of cattle. The absence of the aurochs would allow study of Early Neolithic cattle without the complication of the potential presence of the wild bovine in faunal assemblages.

 

Bronze Age

Bronze Age settlement sites have in recent years produced large assemblages of animal bone. The earliest is the Beaker settlement from around the passage tomb at Newgrange, Co. Meath (van Wijngaarden-Bakker 1986). Two large enclosures–Mooghaun, Co. Clare and Haughey’s Fort, Co. Armagh–furnish material dating to around 1100–1000 BC, while Ballyveelish and Lough Gur, Co. Limerick, Chancellorsland, Co. Tipperary, and the stone fort of Dun Aonghasa, Co. Galway, have produced material dating to the first half of the first millennium BC. The data derived from these sites are summarised in Tables 5.4–5.6, while Figure 5.2 provides an overview of the minimum numbers of individuals of the main domesticates.

What is most striking is the very low incidence of wild mammals at any of the sites. The highest frequency of red deer was at Beaker period Newgrange, but even in that instance, deer represent a very low level of exploitation comprising only 0.8% of the fragments total (van Wijngaarden-Bakker 1986, 89). It is presumed that red deer were deliberately reintroduced into Ireland during the Neolithic, but the date at which this occurred cannot as yet be established. It is possible that they were introduced because of the value of antler as a raw material rather than as a source of meat protein. The Newgrange Beaker assemblage also provides the first evidence for the presence of horse in Ireland. Van Wijngaarden-Bakker (1975) has noted that the appearance of the domesticated horse in Western Europe coincides with evidence for Beaker occupation. The potential presence of wild horse, however, can create problems in identifying domesticated horse. Such complications do not exist in Ireland, however, as there is no evidence for horse here since before the last glacial maximum. The few references to horse bones that have been found in megalithic tombs are either probably secondary intrusions or mis-identifications. For example, a horse skull fragment from Audleystown, Co. Down, is likely to be associated with the Food Vessel insertions into the tomb (McCormick 1986, 41). A burnt bone pin from Fourknocks passage tomb, Co. Meath, was described by Hartnett (1957, 245) as having been made of a horse metapodial. Examination of the pin by the writer, however, indicates that it is not possible to identify the bone to species level. As such, the present evidence strongly suggests that the horse was introduced into Ireland at the beginning of the Bronze Age (McCormick 2005, 17–19).

Table 5.4: Distribution (%) of main domesticates from Irish Bronze Age sites after McCarthy (forthcoming a), McCormick (1987), McCormick (1991a) and O’Shaughnessy (forthcoming).

Table 5.5: Fragments and minimum number of individuals (MNI) data from Bronze Age sites (van Wijngaarden-Bakker 1986; Murphy and McCormick 1996; McCormick and Murray 2006). The MNI values from Mooghaun were estimated by combining the MNI values from a large number of relatively small samples but the distribution coincides with the pattern of the largest samples.

Table 5.6: Fragments and minimum number of individuals (MNI) data from Later Bronze Age sites (McCormick 1987; van Wijngaarden-Bakker 1995*; McCarthy forthcoming a; McCormick and Murphy forthcoming; O’Shaughnessy forthcoming**).

Figure 5.2: Minimum numbers of individuals distributions from Irish Bronze Age sites based on the data contained in Table 5.4.

The nature of Bronze Age faunal assemblages indicates that a range of livestock economies were practiced (Table 5.5; Fig. 5.2). In terms of MNI, the animal of principal importance varies greatly, which is perhaps unsurprising in a period that spans nearly two millennia. Pig were dominant at Beaker Newgrange, while sheep/goat predominated at Dún Aonghasa, a cliff-edge site on the island of Inishmore off the coast of County Galway. Cattle were dominant in the two sites from the middle of the period, namely Haughey’s Fort and Mooghaun. In later sites, with the exception of Dún Aonghasa, there tended to be a more balanced livestock economy, with pig and cattle tending to have been of equal importance.

 

Early Bronze Age

The Beaker assemblage from Newgrange lies at the Neolithic/Bronze Age transition and is characterised by a high incidence of pig. It has been noted that in England, the Early Neolithic is characterised by high numbers of cattle, but that pig became increasingly important towards the end of that period (Grigson 1982). This is particularly notable in henge monuments associated with Grooved Ware pottery. The English evidence is more equivocal at the beginning of the Bronze Age but the largest sample, from Grooved Ware/Beaker levels at Mount Pleasant, Dorset, shows a continued predominance of pig. It may well be that the Newgrange material is part of this pattern. Grigson (1982, 309) equates the increase in pig at this time to a decline in pastoralism occasioned by a regeneration of woodland and especially bracken which can accompany such regeneration. Pigs can thrive on bracken, but it is poisonous to cattle, sheep and horse. Does the pollen evidence support such an interpretation? The evidence from Redbog, Co. Louth, supports the occurrence of a regeneration of woodland from about 2300 cal. BC to about 1600 cal. BC, when extensive clearance begins again (Weir 1995). While bracken is continually recorded as being present, there are no peaks compared with preceding periods. The high levels of pig at Newgrange can most likely be explained in the context of increased woodland, with a decline in available pastureland, more suitable for sheep and cattle.

Mount (1994), however, questions the assumption that the faunal assemblage from Newgrange is representative of the livestock economy of the period. He questions the excavator’s interpretation of the settlement as being merely ‘people … squatting in rather flimsy structures’ around the edge of the collapsed cairn of the passage grave (O’Kelly 1989, 73). Instead, Mount regards the faunal material only as an indicator of sporadic ritual activity associated with the Early Bronze Age pit circle excavated by Sweetman (1985). He also regards the English faunal assemblages associated with Late Neolithic henges as being unrepresentative of the contemporary livestock economy. It is quite likely, as Mount argues, that the food refuse was a product of feasting, but there is no reason to believe that the food consumed on such occasions, whether it be ritual or social feasting, should be unrepresentative of the livestock economy in general. It is probable that in most early societies, the ritualistic behavior of feasting is largely a direct consequence of having to deal with the large amount of fresh meat that became available when a domesticated animal was slaughtered (McCormick 2002a).

The pigs reared at Newgrange would have been used exclusively used for their meat. Van Wijngaarden-Bakker (1986, 74–5) found that the highest proportion were killed at about 2–2.5 years, but that a significant number were killed at an older age. She concluded that they were a late maturing type. In the case of cattle and sheep from Newgrange (there was no evidence for goat) there was the potential for the production of secondary products. The identification of secondary exploitation of domesticates can be based on analysis of the sex ratio and age-slaughter patterns, generally using models outlined by Payne (1973). The main problem with this is that faunal samples often do not provide adequate aging and sexing data. Moreover, as Greenfield (2005) emphasises, mixed livestock strategies rather than specialised secondary produce strategies would probably have been the norm in prehistoric societies. As a result it is likely that early sex ratio and age-slaughter patterns might not conform to Payne’s models. There was an equal balance between male and female cattle at Newgrange, with the peak in slaughter occurring in three and four year olds, with only about 10% of the animals surviving into their fifth year. On the basis of this information, van Wijngaarden-Bakker (1986, 48–51) concluded that the animals were primarily being reared for their meat and that there was no evidence for dairying. This need not mean, however, that they were not milked occasionally since recent isotopic analysis of pottery residues has shown that cattle have been milked in England since the Early Neolithic (Copley et al. 2003).

There is no definite evidence for the presence of goat in Ireland by the Beaker period. In her preliminary report on the Newgrange faunal assemblage, van Wijngaarden-Bakker (1974, 77) suggests that two radii could be attributed to goat but withdraws this in her subsequent publication of the assemblage when she stated that ‘no bone could be ascribed with certainty to goat’ (van Wijngaarden-Bakker 1986, 76). The ageing data for sheep at Newgrange was equivocal, suggesting an emphasis on the killing of semi-mature individuals. No data concerning the sex of the individuals is available. The absence of old sheep indicates that the production of wool was unimportant at this time. At present there is no evidence, direct or otherwise, for the exploitation for wool in Ireland before the Late Bronze Age. The earliest apparent spindle whorls are from Late Bronze Age Ballyveelish, Co. Tipperary (Cleary et al. 1987, 25) and Ballinderry I, Co. Meath (Hencken 1942, 19), with the earliest wool textile having been found with the Late Bronze Age Cromagh’s hoard, Co. Antrim (Jørgensen 1992, 19).

 

Late Bronze Age

Two Bronze Age hillfort settlements dating to about 1000 BC have produced faunal assemblages–Mooghaun, Co. Clare and Haughey’s Fort, Co. Armagh. The distribution of the main domesticates at the two sites is quite similar, the main difference being that sheep/goat are more important at Mooghaun (Fig. 5.2). This may be explained by the presence of a variety of land use types in the vicinity of Mooghaun, including more marginal raised bog and fenlands (Grogan 1995, 46), and it is possible that these areas were better suited to sheep rather than cattle grazing. Goat were present at both Mooghaun and Haughey’s Fort but in very small numbers.

The Mooghaun sample was extremely fragmented and provides little information about the age or sex distribution of the livestock present. Cattle ageing, based on epiphyseal fusion, indicated that the majority (67%) of those long bones that fuse at 42–48 months were fused (McCormick and Murray 2006, 305). This compares with a mere 22% in the case of Newgrange, and marks an apparent shift in slaughter pattern to older cattle, although the extreme fragmentation could be producing a bias in the sample. The Haughey’s Fort data originated in waterlogged contexts and should suffer less from taphonomic bias. There, 52% of cattle were in the older age group which again represents a shift towards the exploitation of more mature individuals compared with the situation at Newgrange (Murphy and McCormick 1996, 48). At Haughey’s Fort, a small sample of mandibulae supported the fusion data, with 50% having derived from animals over 40 months at time of death (Murphy and McCormick 1996, 48). The move toward older animals could imply an increasing exploitation of secondary products such as milk. However, Early Christian period age slaughter patterns, which clearly reflect a dairying economy, show a peak in one to two year olds with fewer older animals being present (McCormick 1992). The Later Bronze Age data is therefore not characteristic of a dairying economy. It may well be that these older cattle represent the introduction of traction into Ireland. Van Wijngaarden-Bakker (1986, 96) concluded that there was no evidence for the use of draft oxen at Beaker Newgrange. Furthermore, it is Mitchell and Ryan’s (1997, 234) contention that the plough was introduced into Ireland during the ‘Later Bronze Age’. Unfortunately, there is not enough data from either Mooghaun or Haughey’s Fort to allow the sex ratio of the cattle to be determined.

Because pig are single use species their age slaughter pattern tends to be consistent both chronologically and spatially. At Haughey’s Fort, for instance, the peak in slaughter occurred at 2–2.5 years, a similar pattern to that noted at Newgrange. The small amount of data from Mooghaun seems to confirm this pattern. While the great majority of pigs on these sites were domesticates, small numbers of wild pig were noted at Haughey’s Fort. The hunting of wild animals, at this and all sites of the period, was incidental. Sheep numbers were especially low in Haughey’s Fort where, in terms of MNI, they were less important than dog. The higher incidence of sheep at Mooghaun, as suggested above, may have been due to environmental factors. There was no artifactual evidence for wool processing having occurred at either site.

The horse remains present at Haughey’s Fort were generally derived from old animals, suggesting that they were not slaughtered until they had fulfilled their roles as traction animals. One of the horses at the site displayed spavin which can be caused by heavy traction (Murphy and McCormick 1996, 48). Some metacarpals displayed cut marks that were likely to have been the result of skinning. The breakage of other horse bones may have been for the removal of marrow. The dog population at Haughey’s Fort was characterised by large animals, with shoulder heights of up to 65 cm being recorded. These are amongst the largest known dogs from a prehistoric site in Britain or Ireland at this time (Murphy and McCormick 1996, 49).

In contrast to the preceding period, sites from the latest phase of the Bronze Age display a range of economic strategies as noted at Dún Aonghasa, Co. Galway, Lough Gur, Co. Limerick, and Chancellorsland and Ballyveelish, Co. Tipperary (Table 5.6). The development of wool processing could account for the rise in sheep rearing at this time, although environmental factors must at least partially account for the unprecedented high occurrence at Dún Aonghasa. Cattle do not have the clear MNI dominance noted on the earlier hillfort sites; this relative decline would appear to be related to an increased exploitation of sheep. The enclosed habitation sites of Ballyveelish and Chancellorsland, Co. Tipperary (Doody 1987; 1996), indicate little change in pig levels during the latter part of the Bronze Age (Fig. 5.2). At Lough Gur, pig were the dominant species but not to the same extent as noted at Beaker period Newgrange. Their high incidence is difficult to explain as there is an absence of palynological evidence for the area at this time. Van Wijngaarden-Bakker (1995, 87–9) notes that a bird bone assemblage from a Late Neolithic/Early Bronze site in the same area suggests an open wetland and rough grassland environment with little evidence for forestation. The high incidence of pig, therefore, might suggest re-forestation during the Later Bronze Age but there is no independent evidence for this.

The Dún Aonghasa assemblage is quite different from any of the assemblages so far considered. In virtually all other post-Mesolithic sites wild animals, be they mammals, birds or fish, were either not exploited or were of only incidental importance. In this assemblage, large quantities of fish were present, the species dominated by wrasse and bream (McCarthy forthcoming). The large assemblage of birds present were dominated by guillemot and shag (O’Sullivan forthcoming) and molluscs, characteristic of the rocky shore, were also exploited (O’Connell forthcoming). The mammal assemblage is dominated by sheep, with pigs being found in negligible numbers. Pig constitute about 8% of the MNI total but this is in fact a methodological exaggeration as they constitute only 1% of the identifiable fragments recovered from the site. The low incidence of pig can be attributed to the low incidence of oak on the islands and the fragile nature of the soil cover that would be susceptible to destruction by the rooting habits of pigs (McCormick and Murphy forthcoming). This low incidence of pig is unparalleled in any Irish or British prehistoric site. The high precedence of sheep can usually be attributed to the presence of a strong wool trade, rather than wool production on a subsistence level, or the occurrence of extensive cereal cultivation since their manure provides richer fertilizer than cattle. Both Grant (1984, 543) and Cunliffe (1978, 184) have invoked both of these explanations for the high prevalence of sheep during the southern English Iron Age. It is unlikely, however, that either of these factors can account for the large numbers of sheep at Dún Aonghasa. It seems more probable that the high prevalence of sheep is simply due to the hardiness of the species. Sheep can generally endure more extreme weather conditions than cattle, especially during winter, when they can survive on limited fodder. Indeed, shortage of winter fodder was probably the principal constraint for livestock rearing on Late Bronze Age Arran. Hay was not saved in Ireland until the Anglo-Norman period so livestock would have been dependent on the very meagre vegetation that grew throughout the winter.

Table 5.7: Cattle fusion data from Dún Aonghasa and other Irish prehistoric sites after van Wijngaarden-Bakker (1986), McCormick (1987), Murphy and McCormick (1996), McCormick and Murphy (forthcoming) and O’Shaughnessy (forthcoming). The ageing data is derived from Silver (1969).

Figure 5.3: Cattle age slaughter patterns based on the data contained in Table 5.7.

The marginality of the area is reflected in the age/slaughter patterns of the cattle where there is a high incidence in the slaughter of very young animals. Half of the cattle were slaughtered before the pelvis had fused, which occurs at approximately 7–10 months (Table 5.7; Fig. 5.3). A high juvenile slaughter rate is a feature of coastal sites in the west and north of Scotland for all periods (Noddle 1979), and the Dún Aonghasa material conforms to this pattern. While it can be argued that the slaughter of young animals may reflect a specific livestock rearing regime (Payne 1973), it is more likely in the case of Atlantic sites that the killing of young cattle is a consequence of fodder shortage (McCormick 1998). The slaughter of large numbers of young calves in order to ensure that the remainder had enough fodder for survival is succinctly expressed in the Hebridean proverb ‘Is fearr aon laogh na da chraicionn’–one calf is better then two skins (Carmichael 1916, 256). An unusually high juvenile sheep mortality (there were no goat noted at Dún Aonghasa) was also noted, with 56% of the pelves being unfused (Fig. 5.4).

Cattle age slaughter/patterns from other Late Bronze Age sites do not show consistent results. The high incidence of mature cattle at Ballyveelish is reminiscent of the pattern noted at Haughey’s Fort but the Chancellorsland assemblage, with its high incidence of juvenile animals, has much more in common with slaughter regimes of the Early Christian period.

Figure 5.4: Sheep age slaughter patterns (McCarthy forthcoming a; McCormick and Murphy forthcoming, based on the method of Silver (1969)).

There is ample evidence for the processing of wool during the Late Bronze age after 1000 BC (see above). Despite this, there is still no decisive shift to Payne’s (1973) ageslaughter model for wool production, with its predominance of older animals. The peak in sheep slaughter at Chancellorsland is for semi-mature animals, which according to Payne (1973, 281) would suggest that they were primarily raised for meat (Fig. 5.4).

There is clear regional and temporal variation in livestock size during the Bronze Age. Figures 5.5–5.7 compare cattle, sheep and pig measurements from different sites. In the case of cattle it can be seen that many of the individuals present at Dún Aonghasa were smaller than those on mainland sites (Fig. 5.5). This can be attributed to environmental factors, with the island being less suitable for breeding cattle than other areas. The same seems to holds true for pig (Fig. 5.6), but is not the case with sheep (Fig. 5.7–it is assumed that all the Dún Aonghasa caprovine remains are of sheep). The sheep from Dún Aonghasa are both larger and smaller than those noted from elsewhere, but admittedly the latter samples are very restricted. If the results are accepted as valid, it implies that the environment of Arran was eminently suitable for the raising of sheep.

The cattle from Beaker period Newgrange are extremely large compared with those present on other sites. This finding is unexpected as there appears to have been a general decline in cattle size after domestication. Davis (1987, 178) notes that in England, there is a continual decline in cattle size between the Neolithic and the Iron Age, with Early Bronze Age cattle tending to be smaller than those of the Neolithic. This is not the case in Ireland. The cattle from Newgrange were on average larger than those present in contemporary Late Neolithic/Early Bronze Age Lough Gur and also larger than those from Early Neolithic Ashleypark (McCormick 1997a, 302). The reasons for this dichotomy are at present unclear and more samples from Neolithic sites are needed in order to provide a possible explanation. Metrical data for pigs is somewhat limited for Irish prehistoric sites. In general, the pigs from Newgrange tend to be larger than later periods, with a gradual decline in size occurring between the Early Bronze Age and the Early Christian period.

Figure 5.5: Cattle size on Bronze Age sites based on astralagus measurements (McCormick unpublished data).

Figure 5.6: Pig size on Bronze Age sites based on astralagus measurements (McCormick unpublished data).

Figure 5.7: Sheep size on Bronze Age sites based on astralagus measurements (McCormick unpublished data).

As occurred during the Neolithic, animals continued to be associated with burial ritual during the Bronze Age, but are generally confined to Early Bronze Age single burials. Pig tusks are often included in cist burials, almost invariably in association with Food Vessel pottery (McCormick 1986, 37). There is no clear evidence of joints of meat of a consistent type having been placed with burials, such as the pig shoulder joints that have been noted in some contemporary Scottish burials (McCormick 1991b). Sea shells continue to be deposited with burials. Ritual deposition of animals unassociated with human burials has also been noted; for example, cremated bones of a range of animals were found in the pits of the Early Bronze Age pit circle at Newgrange. The excavator concluded that the pits ‘acted as receptacles for remains of burnt votive offerings’ (Sweetman 1985, 214).

Dog bones too are occasionally found in ritual contexts, a few apparently in association with Early Bronze Age burials (McCormick 1986, 40). The Late Bronze Age artificial pond at the King’s Stables, Co. Armagh, contained a curious assemblage of bone (Lynn 1977). In addition to some bones of the main domesticates, it contained unusually high proportions of dog and red deer antler. The pond also contained parts of a human skull. The author favored a ritualistic deposition interpretation for the assemblage.

Iron Age

The small number of extant Iron Age assemblages are all from high status ceremonial sites. Navan Fort, Co. Armagh (McCormick 1997b), Tara, Co. Meath (McCormick 2002b) and Dún Ailinne, Co. Kildare (Crabtree 1990), are all identified in the early historical literature as regional capitals but the actual functions of the sites are not always clear. The Tara material was derived from the ditch of the large enclosure known as Ráith na Ríg. While the enclosure is of Iron Age date, it enclosed a Neolithic passage grave and two conjoined ‘ringfort’ type monuments, possibly of Early Christian date. The bones consisted of discarded food refuse but it is unclear what sort of habitation activity it reflected. Dún Ailinne was clearly not an ordinary habitation site (Wailes 1990). Morphologically the site is a henge and the large circular and ‘figure of eight’ structures were obviously not dwellings. Wailes (1990, 19) concluded that it was a ceremonial site and the faunal remains may relate to communal activity such as feasting. The Navan material came from the phases that pre-date the massive 40 m ceremonial wooden structure. It is associated with a series of round houses within a palisaded enclosure (Lynn 1997) and the evidence suggests that the occupation was domestic, albeit of high status, as indicated by the artifactual finds present and the occurrence of a Barbary ape skull (Raftery 1994).

The similarity of the sites as regional ceremonial centres is not reflected in the faunal assemblages. Pig clearly predominate at Navan Fort, while cattle are the dominant species at Dún Ailinne (Table 5.8). The samples from Tara were small but suggested domination by cattle. The high incidence of pig at Navan cannot be attributed to environmental suitability–i.e. the presence of large-scale oak forests. The livestock economy of nearby Haughey’s Fort is dominated by cattle and the pollen evidence does not indicate any increase in oak forest cover in the intervening period (Weir 1997, 116). Instead, the high incidence of pig can be interpreted as a reflection of the choice of pork as a high status food during the Iron Age. Strabo noted that pork was the favorite food of the Continental Celts (Tierney 1960, 268), while Ross (1974, 395–6), on the basis of Irish mythology, concluded that pork is the proper food to be served at the feast, in the ritual of hospitality in the courts of kings, and in the dwellings of the gods.

If the assemblage at Navan can be explained in the context of the preferred diet of high status Celts, how does one explain the dominance of cattle at Dún Ailinne? While Navan Phase 3 is clearly a domestic habitation site this is not the case with Dún Ailinne. The large figure of eight structure with its funnel entrance and the subsequent circular structure with its standing or seating platforms suggest great ceremonial constructions for massed open-air assemblies. The scale of these would imply social inclusively rather than restriction to a small number of high status individuals. Feasting in this context would therefore not have been confined exclusively to the elite, but instead include the general population where beef, rather than pork, may have been considered a more appropriate food.

Cattle ageing data is not available for Navan and the material from Dún Ailinne and Tara is limited. It is, however, clear that the patterns observed at Dún Ailinne and Tara differ greatly. The great majority of cattle killed at Tara are mature individuals (McCormick 2002b, 105), while large numbers of young calves were killed at Dún Ailinne (Crabtree 1990, 23). Crabtree argued that the age slaughter pattern at the latter site represented dairying but McCormick (1991c) did not agree with this interpretation. On the basis of the small samples it is difficult to identify specific cattle raising strategies but it is clear that there is regional variation within Ireland. The pig slaughter pattern noted at Navan, the only site to produce an adequate sample, indicates an age slaughter pattern similar to that noted in earlier periods. There is no useful data available for sheep; the species is generally much less important than either cattle or pig during the Irish Iron Age. This contrasts greatly with southern England at this time where sheep were the dominant species on the majority of sites (Hambleton 1999).

Table 5.8: Fragments and minimum numbers of individuals distribution from Iron Age sites (Crabtree 1990; McCormick 1997; McCormick 2002b). NB: The sample from Tara was too small to allow reliable calculation of MNI values.

Although horse was present in Ireland since the Beaker period they occur in very small quantities during the Bronze Age (see above). They are absent from Dún Aonghasa and Chancellorsland and only a few fragments were noted at both Haughey’s Fort and Lough Gur. The high incidence of horse fragments at Ballyveelish was exceptional, but most fragments were teeth and they represented only a single individual from the two main contexts on the site. The incidence of horse increased greatly during the Iron Age, and is perhaps a reflection of the development of wheeled vehicles as evidenced by the Doogarymore wheels which date to the Late Bronze Age/Early Iron Age transition (Raftery 1994, 104). At Dún Ailinne and Tara, horse remains comprise 2.5% and 5.5% of the fragments’ totals respectively. The horse remains are interspersed with other discarded food refuse so it must therefore be concluded that horse flesh was being consumed. Some of the bones from Tara displayed butchery marks and in one instance the bone displayed evidence of roasting. Dog bones from Tara also displayed butchering marks. It is possible that the high incidence of horse bones at Tara may derive from rituals associated with kingship (McCormick 2005, 21–2). The consumption of dog flesh may too have ritualistic associations (Bhreathnach 2002, 118–20; McCormick 2002b, 107).

The complete horse long bones from Tara provide us with the first evidence of horse shoulder heights in Ireland. They are from animals with estimated withers’ heights of 130 cm and 133 cm, i.e. 13–14 hands (McCormick 2002b, 107), comparable in size to the modern Connemara pony.

The association of animal bones with burial ritual continues during the Iron Age but to a lesser extent than previous times. Only a small number of Iron Age burials are known from Ireland, mostly consisting of low mounds or ring barrows. These generally contain cremations in pits either within the mound or under it but inhumations also occur. When animal bones are present, their role in the burial ritual is generally unclear. At Furness, Co. Kildare, the matrix of the mound contained a scattered assemblage of animal bones which included cattle, sheep/goat, pig and horse, together with several disarticulated human bones (Grogan 1984, 305). No animal bones, however, were noted with the pit cremation burials on the site. There are several instances of Iron Age, or possible Iron Age, burials where the association between the faunal remains and the human burials are more likely to be accidental (McCormick 1986, 46; Buckley et al. 2002). It can be concluded that animal deposition did not play an important role in burial ritual at this time and indeed with the arrival of Christianity, the inclusion of grave goods of any type became theologically unacceptable, although exceptions could be made in special circumstances. A twelfth century document records that the father of Dermait Mac Murchada was buried with a dog as a mark of ‘hatred and contempt’ (Fry 1999, 107).

Conclusions

Faunal studies undertaken during the last 30 years have greatly increased our understanding of how animals were exploited in prehistoric Ireland. There are large gaps in our knowledge but a framework now exists for prioritising areas of future research. The Irish Mesolithic, uniquely in western Europe, was dominated by the exploitation of wild pig, an animal that had the potential to become domesticated. A genetic study is presently being undertaken to establish if early Irish domestic pigs are derived from the native stock or, instead, were being imported as part of a Neolithic ‘package’. Is the dominance of pig noted at Beaker period Newgrange a continuation of a Neolithic livestock economy that was dominated by pig rearing? Neolithic domestic assemblages are needed to answer this fundamental question. The Bronze Age is characterised by considerable variation in livestock rearing strategies across Ireland. These are likely to reflect localised needs and traditions adaptations and contrast greatly with the Early Christian period where a country-wide livestock economy based on dairying developed. The Iron Age assemblages come from contexts that are probably unrepresentative of the general livestock economies of the period. Crucially, assemblages are needed from the first half of the first millennium AD which will allow us to further investigation the origins of the livestock economy that is so well documented during the Early Christian period.

References

Ashbee, P. 1970. The Earthen Long Barrows of Britain. London: Dent.

Barber, J. 1997. The Excavation of a Stalled Cairn at the Point of Cott, Westray, Orkney (AOC Scotland Monograph 1). Edinburgh: Scottish Trust for Archaeological Research.

Bhreathnach, E. 2002. Observations of the occurrence of dog and horse at Tara. Discovery Programme Reports 6, 117–22.

Buckley, V. M., Buckley, L. and McCormick, F. 2002. A curious inhumation burial from Dernish Island, Co. Sligo, pp. 117–20 in Timony, M. A. (ed.), A Celebration of Sligo: First Essays of the Sligo Field Club. Carrick on Shannon: Sligo Field Club.

Carmichael, A. 1916. Grazing and agrestic customs of the Outer Hebrides. The Celtic Review 10 (1914–1916), 40–54, 144–148, 254–262, 358–375.

Chaix, L., Bridault, A. and Picavet, R. 1997. A tame brown bear (Ursus arctos L.) of the Late Mesolithic from La Grande-Rivoire (Isère, France). Journal of Archaeological Science 24, 1067–74.

Cleary, R. M., Hurley, M. F. and Twohig, E. A. (eds.) 1987. Archaeological Excavations on the Cork-Dublin Gas Pipeline (1981–82) (Cork Archaeological Studies No. 1). Cork: Department of Archaeology, University College Cork.

Collins, A. E. P. 1954. The excavation of a double horned cairn at Audleystown, Co. Down. Ulster Journal of Archaeology 17, 7–56.

Connolly, M. 1999. Discovering the Neolithic in County Kerry. A Passage Grave at Ballycarthy. Bray: Wordwell.

Conwell, A. E. 1873. Discovery of the tomb of Ollamh Fodhla. Dublin: McGlashan and Gill.

Copley, M. S., Berstan, R., Dudd, S. N., Docherty, G., Mukherjee, A. J., Straker, V., Payne, S. and Evershed, R. P. 2003. Direct chemical evidence for widespread dairying in prehistoric Britain. Proceedings of the National Academy of Science 100, 1524–9.

Crabtree, P. 1990. Subsistence and ritual: the faunal remains from Dun Ailinne, Co. Kildare, Ireland. Emania 7, 22–5.

Cunliffe, B. 1978. Iron Age Communities in Britain. London: Routledge.

Davis, O. and Evans, E. E. 1933. Excavation at Goward near Hilltown, Co. Down. Proceedings of the Belfast Natural History and Philosophical Society 1932–33, 90–105.

Davis, S. J. 1987. The Archaeology of Animals. London: Batsford.

Doody, M. 1987. Late Bronze Age settlement, Ballyveelish 2, Co. Tipperary, pp. 22–35 in Cleary, R. M., Hurley, M. F. and Twohig, E. A. (eds.), Archaeological Excavations on the Cork-Dublin Gas Pipeline (1981–82) (Cork Archaeological Studies No. 1). Cork: Department of Archaeology, University College Cork.

Doody, M. 1996. Ballyhoura Hills Project. Discovery Programme Reports 4, 15–25.

Frazer, F. C. and King, J. E. 1954. Faunal remains, pp. 70–95 in Clarke, J. G. D., Excavations at Starr Carr. Cambridge: Cambridge University Press.

Fry, S. L. 1999. Burial in Medieval Ireland 900–1500. Dublin: Four Courts Press.

Grant, A. 1984. Animal husbandry, pp. 496–548 in Cunliffe, B. (ed.), Danebury: An Iron Age Hillfort in Hampshire, Vol. 2, The Excavation 1969–1978, The Finds (Council of British Archaeology Report 52). London: CBA.

Greenfield, H. J. 2005. A reconstruction of the secondary products revolution in south eastern Europe: on the origins and use of domestic animals for milk, wool and traction in the central Balkans, pp. 14–32 in Mulville, J. and Outram, A. (eds.), The Zoorchaeology of Milk and Fats. Oxford: Oxbow.

Grigson, C. 1982. Porridge and pannage: pig husbandry in Neolithic England, pp. 297–314 in Bell, M. and Limber, S. (eds.), Archaeological Aspects of Woodland Ecology (BAR International Series 146). Oxford: British Archaeological Reports.

Grogan, E. 1984. Excavation of an Iron Age Burial at Furness. Journal of the County Kildare Archaeological Society 16, 298–316.

Grogan, E. 1995. North Munster Project. Discovery Programme Reports 2, 45–61.

Hambleton, E. 1999. Animal Husbandry Regimes in Iron Age Britain (BAR British Series 282). Oxford: Archaeopress.

Hartnett, P. J. 1957. Excavations of a passage grave at Fourknocks, Co. Meath. Proceedings of the Royal Irish Academy 58C, 197–277.

Hencken, H. O’N. 1935. A cairn at Poulawack, Co. Clare. Journal of the Royal Society of Antiquaries of Ireland 95, 191–222.

Hencken, H. O’N. 1942. Ballinderry Crannog, No. 2. Proceedings of the Royal Irish Academy 47C, 1–76.

Hunt, J. 1967. Prehistoric burials at Cahirguillamore, Co. Limerick, pp. 20–42 in Rynne, E. (ed.), North Munster Studies: Essays in Commemoration of Monsignor Michael Moloney. Limerick: Thomond Historical Society.

Jope, M. 1954. Animal remains from Clough Castle, pp. 150–6 in Waterman, D. M., Excavations at Clough Castle. Ulster Journal of Archeology 17, 103–63.

Jope, M. 1955. The animal remains, pp. 77–81 in Collins, A. E. P., Excavations in Lough Faughan crannog, Co. Down, 1951–2. Ulster Journal of Archaeology 18, 47–82.

Jørgensen, L. B. 1992. North European Textiles until AD 1000. Aarhus: University Press.

Liversage, G. D. 1968. Excavations at Dalkey Island, Co. Dublin, 1956–59. Proceedings of the Royal Irish Academy 66C, 53–233.

Lynn, C. J. 1977. Trial excavations at the King’s Stables, Tray townland, County Armagh. Ulster Journal of Archaeology 40, 42–62.

Lynn, C. J. (ed.) 1997. Excavations at Navan Fort 1961–71 (Northern Ireland Archaeological Monographs 3). Belfast: HMSO.

Mallory, J. P. 1981. The ritual treatment of horse in the early Kurgan tradition. Journal of Indo-European Studies 9, 205–26.

McCarthy, M. 1999. Faunal remains, pp. 85–92 in Woodman, P. C., Anderson, E. and Finley, N., Excavations at Ferriter’s Cove: 1983–95. Bray: Wordwell.

McCarthy, M. forthcoming a. The animal bones from Chancellorsland, Co. Tipperary.

McCarthy, M. forthcoming b. Report on the birds, fish and sea mammal bones from Dún Aonghasa, Co. Galway.

McCormick, F. 1985. The animal bones, pp. 89–94 in Manning, C., A burial mound at Ashleypark, Co. Tipperary. Proceedings of the Royal Irish Academy 85C, 61–100.

McCormick, F. 1986. Faunal remains from prehistoric Irish burials. Journal of Irish Archaeology 3, 37–48.

McCormick, F. 1987. The animal bones, pp. 26–9 in Doody, M., Late Bronze Age settlement at Ballyveelish 2, Co. Tipperary, pp. 22–35 in Cleary, R. M., Hurley, M. F. and Twohig, E. A. (eds.), Archaeological Excavations on the Cork-Dublin Gas Pipeline (1981–82) (Cork Archaeological Studies No. 1). Cork: Department of Archaeology, University College Cork.

McCormick, F. 1988. The animal bones, pp. 182–4 in Gowan, M., Three Irish Pipelines: New Archaeological Evidence in Munster. Bray: Wordwell.

McCormick, F. 1991a. The animal bones from Haughey’s Fort. Emania 8, 27–33.

McCormick, F. 1991b. Animal bones, pp. 113–14 in Dalland, M., A short cist at Grainfoot, Longniddry. Proceedings of the Society of Antiquaries of Scotland 121, 111–15.

McCormick, F. 1991c. Evidence for Dairying at Dun Ailinne? Emania 8, 57–9.

McCormick, F. 1992. Early faunal evidence for dairying. Oxford Journal of Archaeology 11, 201–9.

McCormick, F. 1997a. Animal Bone analysis from Neolithic and Beaker complexes, pp. 301–6 in Eogan, G. and Roche, H., Excavations at Knowth (vol. 2). Dublin: Royal Irish Academy.

McCormick, F. 1997b. The animal bones from Site B, pp. 117–20 in Lynn, C. J. (ed.), Excavations at Navan 1961–71 (Northern Ireland Archaeological Monographs 3). Belfast: HMSO.

McCormick, F. 1998. Calf slaughter as a response to marginality, pp. 49–51 in Mills, C. M. and Coles, G. (eds.), Life on the Edge: Human Settlement and Marginality (Oxbow Monograph 100). Oxford: Oxbow.

McCormick, F. 1999. Early evidence for wild animals in Ireland, pp. 355–71 in Benecke, N. (ed.), The Holocene History of European Vertebrate Fauna (Archäologie in Eurasien 6). Berlin: Verlag Marie Lindorf.

McCormick, F. 2002a. The distribution of meat in a hierarchical society: the Irish evidence, pp. 25–31 in Miracle, P. and Milner, N. (eds.), Consuming Passions and Patterns of Consumption (McDonald Institute Monographs). Cambridge: McDonald Institute of Archaeological Research.

McCormick, F. 2002b. The animal bones from Tara. Discovery Programme Reports 6, 103–16.

McCormick, F. 2004. Hunting pig in the Late Mesolithic, pp. 1–5 in Roche, H., Grogan, E., Bradley, J., Coles, J. and Raftery, B. (eds.), From Megaliths to Metals, Essays in Honour of George Eogan. Oxford: Oxbow.

McCormick, F. 2005. Archaeology: The horse in early Ireland, pp. 17–29 in McGrath, M. and Griffith, J. C. (eds.), The Irish Draught Horse: A History. Cork: Collins Press.

McCormick, F. and Murphy, E. forthcoming. The mammal bones from Dún Aonghasa, Co. Galway.

McCormick, F. and Murray, E. V. 2006. The mammal bones from Mooghaun, Co. Clare: preliminary report, pp. 303–6 in Grogan, E., The North Munster Project: Volume 1. The Later Prehistoric Landscape in South-east Clare. Bray: Wordwell.

Mitchell, F. and Ryan, M. 1997. Reading the Irish Landscape. Dublin: Town House.

Molyneaux, T. 1697. A discourse concerning the large horns frequently found under ground in Ireland, concluding from them that the great American deer call’d a moose, was formerly common in that island: with remarks on some other things natural to that country. Philosophical Transactions of the Royal Society of London, 489–512.

Mount, C. 1994. Aspects of ritual deposition in the Late Neolithic and Beaker Periods at Newgrange, Co. Meath. Proceedings of the Prehistoric Society 60, 433–43.

Murphy, E. and McCormick, F. 1996. The faunal remains from Haughey’s Fort. Emania 14, 47–50.

Noddle, B. 1979. A brief history of domestic animals in the Orkney Islands, Scotland, from the 4th millennium B.C. to the 18th century. Archaeozoology 1, 226–303.

O’Connell, M. forthcoming. Report on the marine mollusca and other shore food from Dún Aonghasa, Co. Galway.

O’Kelly, M. J. 1989. Early Ireland: An Introduction to Irish Prehistory. Cambridge: Cambridge University Press.

O’Shaughnessy, J. forthcoming. The animal bones from Lough Gur, Co. Limerick.

O’Sullivan, T. forthcoming. The bird bones from Dun Aonghasa, Co. Galway.

Payne, S. 1973. Kill-off patterns in sheep and sheep and goats: the mandibles from Aşvan Kale. Anatolian Studies 23, 281–303.

Piggott, S. 1962. Heads and hooves. Antiquity 36, 110–18.

Raftery, B. 1994. Pagan Celtic Ireland. London: Thames and Hudson.

Robertson-Mackay, M. E. 1980. A ‘head and hooves’ burial beneath a round barrow, with other Neolithic and Bronze Age sites on Hemp Knoll, near Avebury, Wiltshire. Proceedings of the Prehistoric Society 46, 123–76.

Roche, G. and Stelfox, A. W. 1936. The animal bones from Ballinderry I, pp. 230–5 in Hencken, H. O’N., Ballinderry Crannog No. 1. Proceedings of the Royal Irish Academy 43C, 103–239.

Ross, A. 1974. Pagan Celtic Britain. London: Routledge.

Schulting, R. J. 1998. Slighting the sea: stable isotope evidence for the transition to farming in northwestern Europe. Documenta Praehistorica 25, 203–18.

Schulting, R. 1999. Radiocarbon dates, p. 219 in Woodman, P. C., Anderson, E. and Finley, N., Excavations at Ferriter’s Cove: 1983–95. Bray: Wordwell.

Schulting, R. J. and Richards, M. P. 2002. The wet, the wild and the domesticated: the Mesolithic-Neolithic transition on the west coast of Scotland. European Journal of Archaeology 5, 147–89.

Silver, I. A. 1969. The ageing of domesticated animals, pp. 283–302 in Brothwell, D. and Higgs, E. (eds.), Science in Archaeology (second edition). London: Thames and Hudson.

Stelfox, A. W. 1942. Report on the animal bones from Ballinderry 2 crannog, pp. 67–74 in Hencken, H. O’N., Ballinderry Crannog No. 2, Proceedings of the Royal Irish Academy 47C, 1–76.

Sweetman, P. D. 1985. A late Neolithic/Early Bronze Age pit circle at Newgrange, Co. Meath. Proceedings of the Royal Irish Academy 85C, 195–221.

Tierney, J. J. 1960. The Celtic ethnography of Posidonius. Proceedings of the Royal Irish Academy 60C, 189–275.

Wailes, B. 1990. Dún Ailinne: a summary excavation report. Emania 7, 10–21.

Weir, D. A. 1995. A palynological study of landscape and agricultural development in County Louth from the second millennium BC to the first millennium AD. Final report. Discovery Programme Reports 2, 77–126.

Weir, D. A. 1997. An outline of the vegetational history in the Navan area from the late mesolithic to medieval times, pp. 111–17 in Lynn, C. J. (ed.), Excavations at Navan 1961–71 (Northern Ireland Archaeological Monographs 3). Belfast: HMSO.

Wijngaarden-Bakker, L. H. van 1974. The animal remains from the Beaker settlement at Newgrange, Co. Meath: first report. Proceedings of the Royal Irish Academy 74C, 313–83.

Wijngaarden-Bakker, L. H. van 1975. Horse in the Dutch Neolithic, pp. 341–4 in Clason, A. T. (ed.), Archaeozoological Studies. Amsterdam: Elsevier.

Wijngaarden-Bakker, L. H. van 1985. The faunal remains, pp. 71–6 in Woodman, P. C., Excavations at Mount Sandel 1973–77. Belfast: HMSO.

Wijngaarden-Bakker, L. H. van 1986. The animal remains from the Beaker settlement at Newgrange, Co. Meath: final report. Proceedings of the Royal Irish Academy 86C, 17–111.

Wijngaarden-Bakker, L. H. van 1989. Faunal remains and the Irish Mesolithic, pp. 125–33 in C. Bonsall, C. (ed.), The Mesolithic in Europe. Edinburgh: John Donald.

Wijngaarden-Bakker, L. H. van 1995. Subsistence in the Late Bronze Age at Lough Gur, pp. 79–92 in Cleary, R., Bronze Age settlement and prehistoric burials, Lough Gur, Co. Limerick. Proceedings of the Royal Irish Academy 95C, 1–92.

Wilde, W. 1840. On the animals and antiquities recently found in Dunshaughlin. Proceedings of the Royal Irish Academy 1, 420–6.

Wilde, W. 1857–61. On the ancient and modern races of oxen in Ireland. Proceedings of the Royal Irish Academy 7, 64–75.

Woodman, P., McCarthy, M. and Monaghan, N. 1997. The Irish Quaternary project. Quaternary Science Reviews 16, 129–59.

Woodman, P. C., Anderson, E. and Finley, N. 1999. Excavations at Ferriter’s Cove: 1983–95. Bray: Wordwell.

Zvelebil, M. 1995. Hunting, gathering, or husbandry. Management of food resources by the late Mesolithic communities of temperate Europe. MASCA Research Papers in Science and Archaeology 12, 79–104.