Journal of the North Atlantic Dairy Farmers and Seal Hunters: Subsistence on a Norse Farm in the Eastern Settlement, Greenland Georg Nyegaard No. 37 2018 An archaeology and environmental history journal focusing on the peoples of the North Atlantic, their expansion into the region over time, and their interactions with their changing environments. Journal of the North Atlantic Board of Editors Símun V. Arge, , Faroe Islands Jette Arneborg, Denmark Colleen Batey, Scotland, UK Gerald F. Bigelow, USA Steven A. Birch, Scotland, UK Colin Breen, Northern Ireland Mike J. Church, England, UK Christyann Darwent, USA Jane Downes, Scotland, UK Andrew J. Dugmore, Scotland, UK Kevin J. Edwards, Scotland, UK Mark Gardiner, Northern Ireland, UK Erika Guttmann-Bond, The Netherlands Agnar Helgason, Iceland Joerg-Henner Lotze, USA, Publisher Niels Lynnerup, Denmark Thomas H. McGovern, USA Jacqui A. Mulville, Wales, UK Anthony Newton, Editor Georg Nyegaard, Greenland Ulla Odgaard, Denmark Astrid E.J. Ogilvie, USA Tadhg O'Keeffe, Ireland Bjørnar Olsen, Norway Richard D. Oram, Scotland, UK Michael Parker-Pearson, England, UK Else Roesdahl, Denmark Alexandra Sanmark, England, UK Niall Sharples, Wales, UK Ian A. Simpson, Scotland, UK, Przemyslaw Urbanczyk, Poland Orri Vésteinsson, Iceland Alex Woolf, Scotland, UK James Woollett, Canada The Journal of the North Atlantic (JONA) is a multi-disciplinary, peer-reviewed and edited archaeology and environmental history journal focusing on the peoples of the North Atlantic, their expansion into the region over time, and their interactions with their changing environments. The journal—published online in the BioOne.org da- tabase and on the JONA website, and indexed in a full range of journal databases—serves as a forum for researchers, and as an information resource for instructors, students, and the intellectually curious who would like to learn about the latest research and study opportunities within the region. The journal publishes a wide diversity of research papers, as well as research summaries and general interest articles in closely related disciplines, which, when considered together, help contribute to a comprehensive multi-disciplinary understanding of the historical in- terplay between cultural and environmental changes in the North Atlantic world. Specifically, the journal’s focus includes paleo-envi- ronmental reconstruction and modelling, historical ecology, archae- ology, ecology of organisms important to humans, anthropology, hu- man/environment/climate interactions, climate history, ethnography, ethnohistory, historical analyses, discussions of cultural heritage, and place-name studies. The journal publishes individual papers on an article-by-article basis. Whenever a manuscript has completed its peer review process and the article galley has been approved by the author, it will be im- mediately published online in the BioOne database and on the JONA website. This publishing model is also available for special volumes such as conference and symposium proceedings or other collections of papers. In effect, this means that articles are grouped online over time, i.e., the table of contents of volumes will grow as articles are posted online, which has the advantage of rewarding prompt authors, while enabling tardier authors to retain the option of being included in a special volume without delaying its publication. The Journal of the North Atlantic’s publishing format is versa- tile enough that authors can include supplementary files with their articles. These supplements may include dataset, figure, and table files (e.g., files requiring a larger than normal journal page size, such as large maps), as well as text and protocol files, and audio and video files (e.g., for ethnographic studies). The Journal of the North Atlantic (Online ISSN #1935-1933, Print ISSN #1935-1984), with an international editorial board, is a collaborative publishing effort of the Eagle Hill Institute, PO Box 9, 59 Eagle Hill Road, Steuben, ME 04680- 0009 USA. Phone 207-546-2821, FAX 207-546-3042. E-mail: office@eaglehill.us. Website: www.eaglehill.us/jona. Copyright © 2018, all rights reserved. On-line secure subscription ordering: rate per year is $40 for individuals, $32 for students, $250 for organizations. Authors: Instructions for authors are available at www.eaglehill.us/jona. The Eagle Hill Institute (Federal ID # 010379899) is a tax exempt 501(c)(3) nonprofit corporation of the State of Maine, USA. Cover Image: A wooden toy horse found in the mire (L = 7 cm). It shows an unmistakable resemblance to the Icelandic breed. Photo by Geert Brovad of the Zoological Museum, University of Copenhagen, Copenhagen, Denmark. “Skálholt Map” courtesy of The Royal Library, Copenhagen, Denmark 2018 Journal of the North Atlantic No. 37 2018 Journal of the North Atlantic No. 37:1–80 G. Nyegaard Dairy Farmers and Seal Hunters: Subsistence on a Norse Farm in the Eastern Settlement, Greenland Georg Nyegaard* Abstract – The paper presents the results of the zooarchaeological investigation of a large bone sample from a stratified bog deposit situated next to the dwelling of a Norse farm in South Greenland, ca. 6 km NNW of the presumed political centre of the Landnam period in the Eastern Settlement. The bog layers containing cultural remains cover a period of at least 300 years. The stratigraphical analyses show no dramatic changes in species composition through time, indicating a rather conservative economic system. The age structure of slaughtered cattle reflects a focus on a dairy economy. Investigations of the bones of the small ruminants show that almost equal numbers of lambs and kids were born at the farm, but that the adult population contained twice as many sheep as goats. Pigs were eaten occasionally, but no bones from adults have been found to document the presence of a local breeding stock. Finds of several bones of cat add this species for the first time to the domestic stock in Norse Greenland. Eleven species of wild mammals make up a little less than half of all identified bones of mammals. The seals dominate and the majority are from harp seal and hooded seal. Introduction In 1997–98 Qaqortoq Museum carried out a res­ cue excavation of mire deposits containing well pre­ served Norse remains at ruin group Ø34 in the Qor­ lortoq valley in South Greenland prompted by the planned expansion of fields. The valley of Qorlortup Itinnera constitutes the northern limit of the lowland area located between the inner parts of the two South Greenlandic fjords of Tunulliarfik and Nordre Ser­ milik; this area hosts the greatest concentration of farmsteads, both medieval and from modern times (Fig. 1). Ca. 6 km SSE of the ruin group is todays’s settlement of Qassiarsuk, which has been identified as Erik the Red’s Brattahlið, the place where this pioneer of the medieval Nordic landnam chose to settle. In Qorlortup Itinnera there are the ruins of 7 Norse farmsteads which, with a point of departure in Tunulliarfik, comprise ruin groups Ø33, Ø34, Ø35, Ø36, Ø37, Ø38, and Ø4, according to Daniel Bruun’s numbering system (Fig. 1) (Guldager et al. 2002, Krogh 1982). In addition, in the mountains to the north and south of the valley, there are respect­ ively 6 and 3 small groups of ruins, interpreted as shielings, which have had functional links with the farms in the valley. Ruin group Ø34 lies in fertile surroundings ca. 3 km from Tunulliarfik (Fig. 2). It consists of 17 ruins and structures belonging to a small or medium sized farmstead as shown on the location plan (Fig. 3). The central part of the ruin complex (ruin nos. 3–9 and 13) is located on dry moraine formations running in the direction of the valley. The dwelling (no. 4) is ca. 30 m long and 10 m wide and comprises several adjoining rooms. In continuation, towards the NE, there is a 14 m long byre (no. 3), which apparently functioned as a cowshed (Guldager et al. 2002). ESE of the dwelling and cattle byre, between the foot of the mountain slope and the high ground on which the ruins stand, lies the small mire which was the subject of study. This is dry at the surface, but moisture increases with depth. The mire contained peat deposits of up to 1.5 m in depth. Of these, the upper 80–90 cm in the area in front of the dwelling contained discarded refuse, whereas the uppermost 10–20 cm comprised the living turf layer. During the occupation of the settlement, there was continual peat accumulation such that the mire contained an uninterrupted stratigraphy comprised of successive layers of well-preserved fibrous fen and moss peat with incorporated cultural remains. In addition to thousands of animal bones there were numerous oth­ er finds: wooden tools and implements, children’s toys, wood-working waste, twigs and branches from the floor of the dwelling or its roof construction, tools and sherds of soapstone, worked objects of bone, antler and tusk, textile remains, pieces of skin and hide, burnt/fire-shattered stones, hearth clean­ ings from the dwelling, i.e., a very mixed content of refuse and artefacts originating from both indoor and outdoor activities. The cultural deposits could be followed horizontally for a distance of a 40 m along a NNE–SSW—oriented drainage ditch (Fig. 4)—cut by the local sheep farmer Karl Kleist in 1993. At right angles to the drainage ditch it could be established that, in several places, the cultural remains dwindled to nothing about 10–12 m out into the mire. *Former deputy at the Greenland National Museum and Archives, Hans Egedesvej 8, Postboks 145, 3900 Nuuk, Greenland. georg.nyegaard@gmail.com. 1 2018 Journal of the North Atlantic No. 37 G. Nyegaard Figure 1. Map of the central part of the “Eastern Settle­ ment” showing all known Norse ruin groups in the area. From Guldager et al. 2002. 2 2018 Journal of the North Atlantic No. 37 G. Nyegaard Figure 2. Overview of ruin group Ø 34 in the Qorlortoq valley. Figure 3. Plan of the central part of ruin group Ø34 by Knud J. Krogh, showing the locations of the excavation trenches in 1997-1998. The ruins are: 3. Probable cattle byre, c. 15 x 6 m. 4. Dwelling, c. 30 x 10 m. 7. Ruin of house, c. 10 x 3 m. 8. Ruin of stone-built house, 8 x 5 m. 9. Ruin of stone-built house, 8 x 5 m. 10. Stone-built pen, c. 10 x 5 m. 11. Stone-built pen, c. 7 x 7 m. 12. Probable house ruin. 3 2018 Journal of the North Atlantic No. 37 G. Nyegaard The Archaeological Excavations1 A total area of 70 m2 was excavated (Figs. 3, 5, and 6). In 1997, 4 trenches— A, B, C, and D—each measuring 2 x 3 m, were excavated. Trench A was located on dry ground to the west of the drainage ditch dug by the sheep farmer—part of an activity area between the dwelling and the mire. Trenches B, C and D were located to the east of the drainage ditch—out in the mire. In 1998, the excavation was expanded with the cutting of trenches E, F, G, H, I, K, L, M, N, O, P, Q, and S. The 1 m wide trenches F and N represent an extension of trench A, running up to the foot of the higher ground which was the site of the ruined dwelling. The other trenches, with the exception of the small trench L, were located out in the mire (Fig. 7). The approach employed was the same each year. Following removal of the grass turf, animal bones were recovered and recorded according to individual 10 cm vertical spits ((layer 0–10, layer 10–20, layer 20–30, etc.), in each square metre. Sieving was not employed as this would have required wet process­ ing of the fibrous peat—a very difficult operation to carry out on site. Another important factor was that wet-sieving would have resulted in significant pollution of the small river running along the val­ ley floor, and of its population of arctic char. On the other hand, conditions were very favourable for the detection and recovery of finds in situ and for observing even very small bone fragments during excavation, notably in the wet deposits where the Figure 5. Plan of the area excavated in 1997-98 on both sides of the modern drain ditch. The ruins of the dwelling are in the white area to the left. Figure 4. The drain ditch dug by sheep farmer Karl Kleist in 1993. Photo: Joel Berglund. 4 2018 Journal of the North Atlantic No. 37 G. Nyegaard bones were conspicuous against the surrounding peat. Furthermore, excavation was generally car­ ried out extremely carefully, and this approach is reflected in the relatively low average weight of 4.2 g for all bone fragments recovered from the excava­ tions in 1997 and 1998. The recorded sections through the mire and the settlement area located between the dwelling remains and the mire are shown on the site plan (Fig. 6). In the main section (Figs. 8A–8E), series of samples were taken from the mire deposits, par­ ticularly in the form of coherent column samples collected for the analysis of pollen, plant macro­ remains and insect remains, radiocarbon dating and other analyses. In 2004, these investigations were supplemented by a series of core samples taken in the middle of the mire, less than 50 m SSE of the dwelling and byre ruins. Sub-samples were then taken for pollen analysis (vegetation history) and AMS radiocarbon dating of plant macro-remains (Schofield et al. 2008). The mire stratigraphy2 In general, sections through the mire reveal a bi­ partite stratigraphy. The lower part is represented by layer 18 (layers 17–18) in the main profile from (x, y) = (105.20, 102) to (x, y) = (109, 102) (Fig. 8D). Figure 7. From a teaching excavation for high school students during the final week of the excavation in 1998 (See Note 1). Figure 6. Trenches excavated in 1997 and 1998. 5 2018 Journal of the North Atlantic No. 37 G. Nyegaard Figure 8A. Section x, y = 93, 102 – x, y = 96.25, 102. Figure 8B. Section x, y = 96, 102 – x, y = 100, 102. 6 2018 Journal of the North Atlantic No. 37 G. Nyegaard Figure 8C. Section x, y = 100, 102 – x, y = 103.70, 102. Figure 8D. Section x, y = 105.2, 102 – x, y = 109, 102. 7 2018 Journal of the North Atlantic No. 37 G. Nyegaard This is a medium- to dark-brown layer of very fine, well-humified fen peat from which cultural remains are completely absent. The upper part comprises alternating layers of brown to yellowish­brown, less humified fen and moss peat, with grass turf up­ permost (layers 1–16). Individual layers could be distinguished on the basis of relative differences in colour, degree of humification, structure and minerogenic content (sand and silt), as well as their content of plant, animal and cultural remains. As can be seen from the section drawings, many of these layers form a coherent unbroken stratigraphy Six conventional radiocarbon dates obtained for peat samples taken from the main section (Fig. 8D) date the mire’s stratigraphy from bottom to top. The dates were obtained for so-called “bulk samples”, i.e., samples from which the coarser roots and cultural remains had been carefully removed. Cali- bration of the radiocarbon dates, marked by the letter K, was carried out using the program Calib 6.0html (Calib 2012, Stuiver and Reimer 1993). A date from the base of layer 18 shows that peat accumulation began at 5708–5490 BC (cal. ± 2 standard devia­ tions) (K-7045). The top of this layer has been dated to 2128–1773 BC (cal. ± 2 standard deviations) (K- 7046). In a core sample from the middle of the mire, the top of the same layer gave a date of AD 240–410 (cal. ± 2 standard deviations) (SUERC-8908). The latter is an AMS date for peat (humic acid fraction) together with selected plant remains. Schofield et al. (2008) interpret this “irregularity” in date for the top of the lower peat layer as resulting from the first inhabitants of the site having cut peat from the mire to be used as building material and for fuel. In both sections a subsequent hiatus is apparent in the sedimentation between the lower humified fen peat (layer 18) and the overlying peat deposits (layer 16). Layer 16 is, like layer 18, a striking contiguous layer in the mire, comprising coarse, light brown, moss­ rich fen peat. The beginning of this layer has more or less synchronous dates in the main section (AD 895–1150 (cal. ± 2 standard deviations) (K-7047)) and in the core sample taken in the middle of the mire (AD 1020–1190 (cal. ± 2 standard deviation) (SUERC-8907)). Cultural remains were found spo­ radically in this layer almost to its base, predomi­ nantly in the form of animal bones. All the overlying layers of moss peat and fen peat contain cultural remains in varying abundance, with the exception of the upper part of layer 1, the living turf layer. In some periods, peat accumulation was dominant and the content of cultural remains modest. In other periods, addition of waste and re­ fuse from the farm dominated. Layers 12 and 10 are thin contiguous moss layers which, in places, have totally retained their green colour. Layer 8 has the greatest concentration of cultural remains. Layer 2 and the lower part of layer 1 are characterised by a large content of charcoal. This could possibly be due to the fact that the farm was burned down when it was abandoned. The next two radiocarbon dates reveal that the most intense occupation of the farm took place from AD 995–1155 (K-7162) to 1035–1251 (K- 7049), whereas the uppermost radiocarbon date in the section dates the end of the occupation to AD 1266–1392 (K-7050) (all three dates calibrated at ± 2 standard deviations). Figure 8E. Section x, y = 109, 102 – x, y = 111, 102. 8 2018 Journal of the North Atlantic No. 37 G. Nyegaard The actual appearance of the mire surface during the occupation is important with respect to an eval­ uation of the depositional conditions and formation processes relating to the farm’s kitchen and butcher­ ing waste. It can be established that, in general, there was not enough open water in the mire basin for gyttja to be formed as this is only seen in the lowermost 5 cm of layer 16. On the other hand, the many fine, coherent and sharply­delimited thin layers of moss show that the mire was very wet in some periods. The excellent conditions for the preservation of organic remains suggest de facto that most of the refuse was thrown out onto a wet mire surface. Furthermore, only 1.8 % of the identified animal bones show traces of gnawing. This must be a consequence of the fact that the mire surface was generally so wet in summer that it did not, for example, function as a common activity surface for the farm’s dogs and pigs. Frequent activity of the latter would also have destroyed or disrupted the clear stratification of the layers. Neither the recorded sections shown in Figures 8A–8E, nor the section recorded between trenches D and P, show any signs of peat cutting, so this ac­ tivity appears primarily to have taken place in the central part of the excavated area when the farm was being established. In the NE part of the mire (trenches E and M) there are clear indications of later digging activity, both in relation to peat cutting and in connection with the creation of a ca. 80 cm wide ditch. The latter runs almost parallel with the modern drainage ditch and cuts through the NW half of trench E and continues into trench M (Figs. 9–10). Like its modern counterpart, this ditch, which Figure 9. Photo of a Norse ditch filled with stones, soil and refuse, sectioned in trench Ea (see Fig. 10). Figure 10. Section x, y = 105.40, 109 – x, y = 108, 109. 9 2018 Journal of the North Atlantic No. 37 G. Nyegaard was later back-filled with stones and refuse, could have been dug in an attempt to drain the mire for cultivation. A cat bone from the ditch fill has been AMS radiocarbon dated to 850 ± 40 BP and reser­ voir corrected to 600 ± 40 BP (AAR-5130). Using the program Calib 6.0html (Calib 2012, Stuiver and Reimer 1993) the calibrated date for the latter is AD 1294–1411 (± 2 standard deviations). Pollen analysis of samples from the core sample taken in the middle of the mire show that Sphagnum sp. (bog mosses) growth increased after the farm was abandoned and that the mire surface became wetter from ca. AD 1420–1630 (SUERC-8905). For information on the vegetation history, see Schofield et al. (2008). AMS radiocarbon dating of animal bones There are several AMS radiocarbon dates for bones from the site, in particular from square 109/102 (Nelson 2008, Takahashi and Nelson 1999). Table 1 shows radiocarbon dates for non­marine animal bones. Lab. nos. E34-86, E34-93 and E34-90 refer to bones excavated during a teaching excavation in 2001. Calibration was performed using the program Calib 6.0htlm (Calib 2012, Stuiver and Reimer 1993). First it must be stressed that Erle Nelson did not collect bones for dating from square 109/102 with the overall chronology of the site in mind. Collectively, the dates suggest a deposition time of ca. 200–250 radiocarbon years. When converted into calendar years, this corresponds to a period extending from some time in the 11th century until the second half of the 13th century. As Takahashi and Nelson (1999) and Nelson (2008) point out, there is no strict correlation between age and depth in square 109/102; for example the oldest of the bones (CAMS-54420269) was recovered roughly from the middle of the deposits. Whereas this cattle astra­ galus could originate from an earlier floor or midden layer which was cleared out at a later time, it is more Table 1. Non-marine radiocarbon dates for animal bones from Ø34. difficult to explain why a few bones found relatively low down in the stratigraphy have yielded rather late dates (i.e., lab. nos. E34-28). The dates from square 109/102 indicate that the stratigraphy could have been disturbed in parts of the bog. What was perceived as a uniform and com­ plete stratigraphy throughout most of the excavation trenches is perhaps in truth not so complete in all places. Even though this is not apparent from the recorded sections, the possibility cannot be excluded that a transverse section at right angles to the main section would have revealed some discontinuities in the layers as a consequence of sporadic horizontal removal of peat. As already mentioned above, there is an AMS radiocarbon date for the left femur of a domestic cat which was recovered from the fill of the ditch that cuts through trench Ea and continues into trench M (AD 1294–1411). This ditch is presumed to have been dug towards the end of the occupation and subsequently back-filled with refuse (Christensen 2000). Dating of the finds: Conclusions relating to the chronology The earliest cultural remains in the mire are represented by scattered animal bones in the light­ brown fibrous fen peat, layer 16 in the main section (Fig. 8D), the base of which is dated to AD 895– 1150 cal. The bones occur almost all the way down to the base of the layer and must originate from an occupation which was contemporaneous with its ac­ cumulation. This conclusion is also supported by the results of the pollen studies which indicate landnam activity at the site from the beginning of the 11th century. The first inhabitants could then have been children or grandchildren of the first immigrants to establish themselves along the shore of the inner part of Tunulliarfik at the end of the 10th century. The radiocarbon dates indicate that the farmstead was abandoned at some time during the 14th century, Sample no. CAMS-54416249 CAMS-54417258 Lab. no. E34-9 CAMS-54420269 Lab. no. E34-28 CAMS-54423276 Lab. no. E34-64 Lab. no. E34-39 Lab. no. E34-72 CAMS-54426288 Lab. nr. E34-86 Lab. nr. E34-93 Lab. nr. E34-2 Lab. nr. E34-90 Square Level 109/102 249 109/102 258 109/102 259 109/102 269 109/102 274 109/102 276 109/102 276 109/102 277 109/102 278 109/102 288 106/94 248 107/94 ? 106/99 282 106/110 235 Layer 10–20 20–30 20–30 30–40 40–50 40–50 40–50 40–50 40–50 50–60 10–20 50–60 0–10 Species Bos t. Bos t. Bos t. Bos t. Bos t. Bos t. Rangifer t. Bos t. Ovis a./Capra h. Ovis a./Capra h. Bos t. Bos t. Bos t. Bos t. Element Radius Radius Calvarium Astragalus Ulna Phalanx 2 Scapula Astragalus Vertebra t. Phalanx 2 Fragment Metatarsus Metacarpus Metatarsus Radiocarbon Age Calibrated age (2 sigma) 770±30 AD 830±40 AD 890±40 AD 990±40 AD 760±40 AD 850±30 AD 940±50 AD 900±40 AD 970±40 AD 860±40 AD 775±30 AD 795±30 AD 900±40 AD 835±25 AD 1217–1281 1052–1274 1035–1219 986–1155 1189–1294 1052–1261 1018–1209 1034–1214 995–1159 1044–1261 1216–1280 1186–1277 1034–1214 1164–1258 10 2018 Journal of the North Atlantic No. 37 G. Nyegaard and this conclusion is supported by the pollen data. Although no find-free horizons were encountered in the mire, it is not possible to determine securely whether there were 250–300 years of continuous occupation or whether breaks occurred. With respect to intra site comparison of finds from the various trenches, it is important to take account of the fact that there are differences in level between the constituent layers in the mire such that the mechanical excavation layers in the individual trenches cannot be compared directly in chronolog­ ical terms. Accordingly, the transition between the lower peat deposits (layers 17–18) and the fibrous, less well-humified peat above (layer 16) is located at different depths below the surface in different places—in the main section: 90–100 cm, in trench E: 65–90 cm, and in the core from the middle of the mire: 54 cm. Material and Methods Bone recovery and state of preservation Even though conditions in the mire were acidic, the cool, constantly waterlogged conditions resulted nevertheless in generally good conditions for the preservation of animal bones, leather, textiles, wood, etc. There is no permafrost in the area, but in waterlogged sediments the ground first becomes completely frost-free in late summer at best. The animal bones had mostly a dark brown patina and were of a tough consistency. Those recovered from the driest excavation layer, layer 0–10, were generally very spongy and porous. The slightly poorer conditions for preservation in this layer see expression in an over­representation of teeth and other resistant bone elements (see below). The bones recovered from the lowermost find- bearing peat deposits, i.e., main section of layer 16, had a lighter reddish­brown colour corresponding to the colour of the peat. After drying, most of the animal bones became noticeably lighter in weight, presumably as a consequence of part of their min­ eral content having been leached out in the mire. Several finds of the membrane-like organic layer that covers the outer surface of mussel shells, the periostracum, bear witness to the dissolution of the calcium carbonate of the shells. Table 2. Cattle metacarpals. Measurements after von den Driesch 1976. Two cattle metacarpals selected for isotope analysis were measured shortly after recovery in 1998 and again several years later when they had completely dried out (Table 2). These measurements show that desiccation lead to a loss of volume which sees clearest expression in the width, depth and, to a lesser degree, the length of these bones. Eleven animal bones were selected for AMS ra­ diocarbon dating from square 109/102; all contained well–preserved collagen. “All samples yielded a white, sponge-like product, as we often find for the extracts of collagen obtained using these methods on bone that is in good condition” (Takahashi and Nelson 1999:3). Extraction of collagen gave yields of 10–27 %; these are typical values for bones that are well-preserved. Some values actually slightly exceeded that expected for fresh bone tissue, but this is considered to be a result of the bones’ inorganic content having been leached out due the acid envi­ ronment in the mire. Eight fish bones recovered from the excavations in 1997 and 1998, of which 3 were found in situ, are all poorly preserved. They had more or less broken down into the basic lamella-like fish bone structure and were consequently very fragile. It is very likely that wet sieving in the field would have led to further fragmentation of any fish bones. A combination of the acid conditions in the mire and alternate freezing and thawing in a waterlogged environment possibly explains their poor preservation. Analysis of the animal bones Following excavation, the animal bones were sent to the Zoological Museum in Copenhagen where they subsequently were examined by the au­ thor. Bones recovered from the spoil heaps alongside the modern ditch in 1994 and during a preliminary investigation in 1995 were examined in 1995. Out of a total of 9013 bone fragments (68.4 kg), in all 3637 (53.2 kg) were identified (Nyegaard 1996a). Only loose teeth and jaw elements of domestic animals and measurable bone elements from this assemblage were included in the present investigation. Animal bones recovered in 1997 and 1998 were examined over a total of 8 months from 2001 to 2007. The resulting data were entered into the database PARA­ DOX. Together with other information, each identi­ Find no. x1253 (metacarpus sinistra) Find Before drying 188.4 mm 57.3 mm 36.9 mm 60.0 mm no. x1252 (metacarpus dextra) GL Bp SD Bd After drying 181.9 mm 54.0 mm 32.6 mm 53.8 mm Reduction 3.5 % 5.8 % 11.7 % 10.3 % Before drying 172.5 mm 56.5 mm 35.3 mm 57.5 mm After drying 164.9 mm 51.3 mm 29.6 mm 51.2 mm Reduction 4.4 % 9.2 % 16.1 % 11.0 % 11 2018 Journal of the North Atlantic No. 37 G. Nyegaard fied fragment was described using a numerical code which enabled calculation of the minimum number of individuals for a species3. Teeth, jaw components and measurable elements were subsequently sent to the Greenland National Museum and Archives for an investigation of age at slaughter and size of individual by the author. All bones of sheep and goat for which morphological criteria are available with respect to species identi­ fication were similarly put to one side and identified together, element by element. The latter took place at the Zoological Museum. The bones recovered during a six-day teach­ ing excavation in 2001 have not been counted or weighed. Teeth, jaw components and measurable elements were, however, included in the overall ma­ terial subjected to measurement and also employed in identification of age at slaughter. Measurements of bones were carried out accord­ ing to the guidelines specified by von den Driesch (1976). Apart from the two bovine metacarpals men­ tioned above, all bones were measured in a dry state. On the basis of the demonstrated loss of volume of these two bones on drying, certain measurements— for example the width of the bovine long bones—can be expected to be as much as 10–15 % less than that of the original bones. The assemblage is stored, together with the other archaeological animal bone assemblages from Greenland, at the Zoological Museum in Copenha­ gen, where there is also access to the digitalised data. Results of the Identifications Totals for mammals, birds and fish from the 1997 and 1998 excavations A total of 42,573 bone fragments (179.3 kg) were recovered from the two excavations, of which 15,701 fragments (36.9 %) (136.2 kg) have been identified to various taxonomic levels (Table 3). Presentation of the overall content is based on a simple summing of all the identified fragments (NISP) for each spe­ cies or animal group. This summary is augmented by a statement of the total bone weight for each group as a tangible expression of the “volume”. The weight of bones of sheep/goat also includes those identified to species either as sheep or goat. The presence of 18 species of mammal has been demonstrated on the basis of bones. Seven of these are domestic species which together constitute 53.3 % of all the identified mammal bones. The remain­ ing 46.7 % originates from at least 11 different wild mammal species. Four groups of mammals domi­ nate the assemblage: cattle (15.9 % of all identified mammal bones), sheep and goat (combined total of 36.4 %), seal (combined total of 39.2 %) and rein­ deer (6.3 %). None of the other species represented exceeds 0.5 %. The totals in the fragment summary do not reflect the actual relationship between the numbers of seals killed of each species. The same applies to the number of sheep slaughtered relative to the corresponding number for goats. In order to draw conclusions in this respect it is necessary to compare the representation of individual anatomical elements of each species. This is done in the section below. Table 3. Summary of the number of identified and unidentified bone fragments recovered in 1997 and 1998. Fragments (NISP) Dog, Canis familiaris Cat, Felis catus Pig, Sus domesticus Sheep, Ovis aries 454 Weight Domestic mammals 30 268 g 21 43 g 76 304 g Goat, Capra hircus Sheep/goat, Ovis aries/Capra hircus 4,829 Cattle, Bos taurus Horse, Equus caballus Wild mammals Arctic hare, Lepus arcticus Arctic fox, Alopex lagopus Polar bear, Ursus maritimus Walrus, Odobenus rosmarus Common seal, Phoca vitulina Ringed seal, Phoca hispida Harp seal, Phoca groenlandica Bearded seal, Erignathus barbatus Hooded seal, Cystophora cristata Seal indet. Reindeer, Rangifer tarandus Beluga, Delphinapterus leucas Beluga/narwhal, Delphinapterus leucas/ 2,453 43,308 g 28 1033 g 11 14 g 23 37 g 19 102 g 48 1,651 g 6 112 g 8 216 g 197 5671 g 15 949 g 268 5793 g 5,562 41,863 g 975 8246 g 2 49 g 6 269 g Monodon monoceros Beluga/narwhal/pilot whale, Delphinapterus l./ Monodon monoceros/Globicephala maelaena 7 Whale (smaller), Cetacea indet. Whale (larger), Cetacea indet. Whale, Cetacea indet. Domestic / Wild mammals Dog/wolf, Canis sp. Unindentified mammal bone fragments Birds Mallard, Anas platyrhynchos White-tailed eagle, Haliaeetus albicilla Rock Ptarmigan, Lagopus mutus Common Guillemot/Brünnich’s Guillemot, Uria aalge/Uria lomvia Razorbill/Common Guillemot/Brünnich’s Guillemot, Alca torda/Uria sp. Unidentified bird bones, Aves sp. Fish 336 g Cod, Gadus morhua 5 Flatfish unspecified, Pleuronectidae 2 Eelpout, Lycodes sp. 1 350 24,127 g 496 g 43 815 g 2 23 g 26,797 43,073 g 1 2g 2 10 g 14 5g 129 155 g 92 105 g 75 18 g 19 3 241 g 12 2018 Journal of the North Atlantic No. 37 G. Nyegaard The majority of the 238 identified bird bones in the assemblage are from either common guillemot or Brünnich’s guillemot. With respect to other species, only the presence of mallard, white­tailed eagle and rock ptarmigan has been securely demonstrated. The fish fauna is represented solely by a few bones of cod, a single bone of eelpout and a couple of bones of unidentified flatfish. Patterns in horizontal distribution A comparison of the horizontal distribution of the species is aided by the fact that the bones are very evenly distributed. Figure 11A shows the num­ ber of bones found in each square metre. The great­ est concentrations, ca. 1100–1700 fragments per square metre, were found in trenches H, P, Ea and the small trench L. Most of the other trenches had 400–1000 fragments per square metre4. The overview of the horizontal distribution is based on a comparison between the four dominant animal groups: cattle, sheep and goat, seal—all spe­ cies combined—and reindeer. Table 4 (on page 18) gives a summary for the various parts of the excava­ tion. (I) is the activity area on dry land between the dwelling and the mire. (II) is from the small trench L located on the edge of the mire. (III) encompasses all the animal bones from the mire, excluding the ditch fill in trench Ea and trench M. (IV) refers to finds from the fill of the ditch which runs through trench Ea and continues into trench M towards the north; a small collection of bones from the southernmost part of square 105/110 is also included here. (V) refers to stray finds while (VI) provides a combined summary for all four animal groups in the total assemblage recovered during the two excavation years. The most striking feature of the horizontal intra site comparison is the very heterogeneous distribu­ tion of reindeer bones (Fig. 11I). These are predomi­ nantly concentrated in the southern half of the exca­ vation in the mire—especially trenches D, H, and P. Here they constitute up to 15–30 % of all identified bones, whereas the proportion seen in the trenches on dry land and in the northern part of the mire is exceptionally modest (0–3 %). A large number of reindeer bones was found within the four squares 106/99, 106/98, 107/99 and 107/98. Away from this concentration their numbers diminish rapidly. The distribution pattern suggests that most of the reindeer bones were deposited over a short period of time and as a result of activity­related and season­ ally-determined circumstances. For example, they could represent refuse thrown directly out from the Figure 11A. Distribution of animal bones in trenches excavated in 1997 and 1998. 13 No. 37 2018 Journal of the North Atlantic G. Nyegaard Figure 11B. Distribution of bones of sheep and goat. Figure 11C. Distribution of identified bones of sheep. 14 No. 37 2018 Journal of the North Atlantic G. Nyegaard Figure 11D. Distribution of identified bones of goat. Figure 11E. Distribution of cattle bones. 15 No. 37 2018 Journal of the North Atlantic G. Nyegaard Figure 11F. Distribution of all seal bones. Figure 11G. Distribution of identified bones of harp seal. 16 2018 Journal of the North Atlantic No. 37 G. Nyegaard Figure 11H. Distribution of identified bones of hooded seal. Figure 11I. Distribution of all reindeer bones. 17 2018 Journal of the North Atlantic No. 37 G. Nyegaard dwelling during winter; this interpretation is con­ sistent with this particular area having the greatest concentration of bone waste. The same four square metres also show a concentration of goat bones: 61 out of the total of 350 identified goat bones originate from here, compared with only 25 identified bones of sheep (Fig. 11C–11D). There is also a small con­ centration of hooded seal bones: a total of 24, com- pared with only 12 of harp seal, which is otherwise the best-represented seal species (Fig. 11G–11H). With respect to the other mammal species, the horizontal distribution is more uniform. Cattle con­ stitute 10–20 % in two thirds of all square metres and slightly more or less than this in the remaining third. There is a slightly higher proportion in the mire than on dry land. There is also a tendency for the cattle to have higher proportions further out in the mire and in the northern half of the large excavation trench. In just less than two thirds of the total excavated area of 70 m2, seal constitute a higher proportion than sheep and goat. In particular, there are high proportions of seal bones in the dry land area between the mire and the remains of the dwelling, whereas a declining trend is seen running from the modern drainage ditch and out into the mire. Accordingly, the proportion of bones of sheep and goat exceeds that for seal bones in a third of the excavated square metres, including part of trenches C, K, and Q in the mire. Patterns in vertical distribution An overview of the distribution of the bones by species, down through the excavation layers, is based similarly on the four animal groups: 1) cattle, 2) sheep and goat, 3) seal and 4) reindeer. The comparison focuses on trenches A, B, C, D, etc. where the layers were dug as coherent surfaces. Summaries for each excavation trench are given in Appendix 1A–D and a detailed survey is presented in Appendix 2, from which the following conclu­ sions have been drawn. In most of the trenches there is a tendency to­ wards a slightly greater representation of cattle in the upper excavation layers. Even though the oppo­ site is also true in trenches L, square meter 103/101, S and Q, which are located alongside the modern drainage ditch, it is possible to draw the significant conclusion that the importance of cattle was not re­ duced through time. Conversely, the slightly higher proportion of cattle bones in the ditch fill in trenches Ea and M (20.8 %), dated to a late phase of the settle­ ment than in the overall summary for the four animal groups (16.2 %) on which the comparison is based, can be seen as showing that the economic signifi­ cance of cattle, in relative terms, increased slightly towards the end of the farm’s lifetime. The overall relative decline in sheep and goat in the upper excavation layers is the most stable trend and is apparent across most of the excavation. The summary for the entire mire (excluding trenches Ea and M) suggests that there was a gradual develop­ ment through a longer sequence of the settlement, whereas the change in the settlement area between the dwelling and the mire first sees concrete expres­ sion in the two uppermost excavation layers. The declining proportion of the small ruminants is accompanied in most cases by an increase in seal bones in the upper layers. In order to evaluate the mutual relative quantities of these two groups “cleaned” of the other species, a sheep/goat:seal ratio has been calculated for each layer in the exca­ vated trenches (Table 5). The lowest values are most commonly seen in the two uppermost layers, with values rising down through the underlying layers. For example, in trench B there is a steady increase from layer 0–10 (0.43) to layer 40–50 (1.22), and there is a fairly even rise through the upper layers in trenches A, F,N,D,H,I,K,Q,R,etc.IntrenchB,theratio does however fall again to 0.1 in layer 50–60. This layer includes the uppermost part of the light brown fibrous fen peat, which is the lowest layer in the stratigraphy containing bones and cultural remains. The same applies to layer 40–50 in trench G, which contains about twice as many bones of seal as of sheep and goat (0.42). A basal layer with low sheep/ goat:seal ratios also occurs in trenches A, H and I. In the overall summary for the mire, excluding trenches Ea and M, gradually increasing ratios are apparent from 0.72 (layer 0–10) to 1.15 (layer 40–50), after which there is a fall to 0.90 (layers 50–60 + 60–70). The figures indicate that in the early phase of the farm, and also in the time towards its conclusion, Table 4. Distribution of identified bones belonging to the four dominant animal groups in different parts of the excavation. I II III IV V VI Settlement area A, F, N Mire B, C, D, Eb, G, H, I, K, O, P, Q, S 1700 15.7 % 4092 37.7 % 4167 38.4 % 883 8.1 % 10842 99.9 % Ditch Ea, M 367 20.8 % 609 34.4 % 746 42.2 % 46 2.6 % 1768 100.0 % Stray finds 55 56 106 11 228 Entire assemblage Trenches Cattle Sheep + goat Seals Reindeer Total 290 778 894 28 1990 14.6 % 39.1 % 44.9 % 1.4 % 100.0 % L 41 14.2 % 98 33.9 % 143 49.5 % 7 2.4 % 289 100.0 % 2453 5633 6056 975 15117 16.2 % 37.3 % 40.1 % 6.4 % 100.0 % 18 2018 Journal of the North Atlantic No. 37 G. Nyegaard there was a period when seal hunting, seen relative to the sheep and goat farming, played a greater role than during the main period of occupation. It should be added that there are also examples of high sheep/ goat:seal ratios in the lowermost excavation layer. The above-mentioned concentration of reindeer bones in the southern part of the excavation trench in the mire is distributed through excavation layers of increasing depth from trench B towards trench D, as is apparent from Table 6, which shows a summa­ ry for every square meter from 107/100 to 107/97. The same tendency is seen in the distribution of the other bones in that the most find­rich layer within these same square metres lies at increasing depth from trench B towards trench D. The levels for the layers containing the greatest number of reindeer bones in squares 107/100 (level 240–260), 107/99 (level 241–263), 107/98 (level 252–271) and 107/97 (level 256–274) do not however deviate much from one another, even though these represent different excavation layers. The small rise in the terrain from trench B towards trench D, a consequence of the more vigorous peat growth in the southern trenches, is one of the explanations for this situation. Fur­ thermore, the fact that the find­rich layers in trench D has a greater thickness than the corresponding layers in trench B also plays a role, as does the fact that the bones were able to sink down further into the wetter part of the mire. The Domestic Fauna (I): Sheep and goat Representation of the two species A total of 5633 fragments derive from the small ruminants, sheep and goat, corresponding to 36.4 % of the identified mammal bones. Of these, 804 frag­ ments (14.3 %) have been identified to species level: 454 sheep and 350 goat. The species identification was based on the works of Boessneck et al. (1964), Gabler (1985), Halstead and Collins (2002), Payne (1969, 1985), and Prum­ mel and Frisch (1986). Some of the morphological differences described by these authors are not so well suited to distinguishing sheep and goat in bone assemblages from Greenland. This is because bone characters in primitive sheep breeds can have a tendency to be “intermediary”, i.e., slightly “goat- like” (Boessneck et al. 1964:61). For example, the conspicuous muscle attachment seen caudally on the collum scapula in many south Scandinavian adult sheep is not, as a rule, so strongly developed in the Norse sheep from Greenland. As an important aid to identification, use was made of bones from a large number of individuals of sheep and goats contained Table 5. Sheep/goat : seal ratios. TrenchA Trench F Trench N K *)A+ F + N 0.31 0.46 1.32 1.10 0.97 1.11 1.07 1.42 *) incl. 103/101 Trench P Trench B 0.43 0.70 0.87 0.82 1.22 *) 0.10 *) incl. 60-70 Trench Qa *) 0.59 0.85 0.47 0.90 **) 1.30 *) incl. 0-10 **) incl. 60-70 Trench C 0.86 1.75 0.96 1.26 1.26 Trench Qb 0.29 1.46 0.97 1.28 *) 1.64 *) incl. 50-60 Top level 230 240 253 263 273 Trench D *) 0.68 0.92 0.80 0.98 1.16 *) incl.- 0-10 Trench S 0.63 0.43 1.05 0.67 2.19 2.40 107/97 NISP 1 0 0 8 33 0 Trench G *) 0.77 2.61 1.27 0.42 *) incl. 0-10 The bog 0.72 0.82 0.90 1.03 1.15 *) 0.90 *) incl. 60-70 Top level 237 246 255 263 273 Layer 0-10 0.47 Layer 10-20 0.19 Layer 20-30 1.29 Layer 30-40 1.37 Layer 40-50 *) 0.65 *) Layer 50-60 Layer 60-70 Layer 70-90 0.20 0.36 0.69 0.46 1.44 1.03 0.97 0.70 1.71 0.69 0.96 1.07 1.42 *)incl. 50-60 *) incl. 50-60 Trench H Trench I Trench Layer 0-10 0.23 Layer 10-20 *) 0.67 0.42 Layer 20-30 0.41 *) 0.80 0.60 Layer 30-40 1.04 1.00 2.89 Layer 40-50 1.26 1.62 1.11 Layer 50-60 0.68 0.78 *) 1.23 0.42 0.53 0.90 1.21 Layer 60-70 *) incl. 0-10 *) incl. 0-10 and 10-20 *) incl. 0-10 Table 6. Vertical distribution of reindeer bones from trench B to trench D. 107/100 107/99 NISP Top level NISP Layer 0­10 0 0 Layer 10­20 25 240 1 Layer 20-30 14 250 12 Layer 30-40 1 260 93 Layer 40-50 1 270 6 Layer 50-60 0 280 0 Top level 229 241 253 261 272 107/98 NISP 0 0 1 61 11 0 19 2018 Journal of the North Atlantic No. 37 G. Nyegaard in the study reference collections at the Zoological Museum, including the skeleton of an entire adult goat found at “Gården under Sandet” in the Western Settlement (Enghoff 2003: 65). An overview of identifications of sheep and goat with respect to individual bone elements provides the opportunity for a more nuanced insight into the composition of the sheep flock and the goat herd (Table 7). Of the total of 804 identified bones, 56.5 % are of sheep and 43.5 % of goat. Fluctuations in the relative abundance of the various body parts evident from Table 7 should be seen in the light of the fact that in most cases it is considerably more difficult to identify post­cranial bones of juvenile animals than of adults (Halstead and Collins 2002). This is par­ ticularly true with respect to most of the long limb bones until the various points in time when the epi­ physes have become fully fused. Conversely, it was possible to identify compact elements, such as the astragalus and calcaneum, to species even in juve­ nile animals. The summary for these bones therefore contains both juvenile and adult individuals. Table 7 shows an equal representation of sheep and goat in the identifications of astragalus and calcaneum. In the case of the former, no distinc­ tion was made between juvenile and adult. This was however done for the calcaneum, in which case the epiphysis, tuberositas calcanei, becomes fused around 3 years of age (Habermehl 1975). If the cal­ caneii are divided up into two age groups, the same quantitative ratio between sheep and goat is seen in the adult animals as in several of the long limb bones, whereas goat comprises by far the majority in the smaller juvenile group. Of the long limb bones, the morphological dif­ ferences between sheep and goat see particular expression in the metapodials. Consequently, the metacarpals and metatarsals identified to species also include juvenile animals. In the overall sum­ mary for all the fragments belonging to these two bones, the relative quantities between sheep (52 %) and goat (48 %) approach the equal representation seen expressed in the astragalus and calcaneum. Species identification on the basis of the pre­ served lower jaw parts with in situ teeth, as well as loose dp4, similarly includes both juvenile and adult animals. In the overall assemblage there is an equal distribution of sheep and goats. On the basis of the investigation it can be con­ cluded that, on average, equal numbers of lambs and kids were born on the farm. The summary for several bones of adult animals, including scapula, pelvis, tibia and calcaneum, reveals a ratio between sheep Table 7. Identification to species of various bone elements of sheep and goat. Cornus(horn cores) Sheep Goat Calvarium Sheep Goat Mandibula, incl. single dp4 Sheep Goat Humerus Sheep Goat Radius, ulna, radius-ulna Sheep Goat Ulna Sheep Goat Tibia Sheep Goat Astragalus Sheep Goat Calcaneum (all bones) Sheep Goat Calcaneum (from adults) Sheep Right Left 20 14 10 12 17 19 19 17 Right/ Left 7 2 40 28 Sum Percentage 41 63% 24 37% 65 40 59% 28 41% 68 36 50% 36 50% 72 Scapula Sheep 9 8 Goat 371037% 17 63% 27 48 61% 31 39% 79 56 56% 44 44% 100 22 63% 13 37% 35 8 10 Goat 26831% 24 24 18 13 27 29 23 21 10 12 6 7 Pelvis Sheep 18 69% 16 16 7 8 17 20 19 19 15 12 15 12 26 32 68% 15 32% 47 37 49% 38 51% 75 27 50% 27 50% 54 13 10 Goat 941336% 36 Sheep 22422% 23 64% Calcaneum (from juveniles) Goat Metacarpus Sheep Goat Metatarsus Sheep Goat Metapodium (epiphyses) Sheep Goat Sum of all metapodials Sheep Goats Sum of all bones Sheep Goat 6 8 14 78% 18 51 56% 40 44% 91 39 53% 34 47% 73 18 17 9 16 14 16 13 9 Distal parts 19 7 17 9 3 3 11 11 14 93 52% 85 48% 178 454 56.5% 350 43.5% 804 20 2018 Journal of the North Atlantic No. 37 G. Nyegaard and goats in the vicinity of 2:1. It can therefore be assumed that twice as many kids and juvenile goats were slaughtered as lambs. Judging from the high proportion of sheep in the identifications of the distal end of the humerus, many goat kids must have been slaughtered at an early age, as the distal epiphysis on this bone fuses already around the age of 3 months (Habermehl 1975). Horizontal and vertical distribution of the identi- fied bones of sheep and goat The sheep and goat bones identified to species have an extensive horizontal distribution with be­ tween 1 and 10 fragments of each species in most square metres (Fig. 11C–11D). Only a few square metres in the northern half of the mire, together with trench N, have a greater number of sheep bones, with values of 12–23 per square metre. A concen­ tration of 61 goat bones in four square metres in trenches H and P coincides, as already mentioned, with the area containing the greatest number of reindeer bones. This could indicate some seasonally determined activities such as slaughtering or the cleaning out of refuse. Table 8 provides a summary of the vertical dis­ tribution of the identified bones of sheep and goat from the various parts of the excavation. As already mentioned, the same layer in different trenches cannot be directly linked chronologically. The relatively few identified bones from the trenches on dry land (A, F, and N) indicate that there were considerably more sheep than goats during most of the occupation, but that the relationship evened out in the later part of the farm’s lifetime. This ten­ dency is especially clear if a comparison is made between the sum for the three upper layers with that for the underlying layers. An overall summary for trenches C, K and Q in the mire reveals the same Table 8. Vertical distribution of bones identified to either sheep or goat. trend, whereby sheep constitute the majority in the lower part of the deposits, with a more equal distri­ bution between the two species in the upper layers. Trenches B, H, P, and S (excl. square meters 105/96 and 105/97) in the mire give another picture. Here, layer 40–50 has equal numbers of bones of sheep and goat and is overlain by layers 30–40 and 20–30 where goat bones are in the majority. A contributory factor to this is, however, that the summary includes the square metres in the mire containing the above­ mentioned concentration of goat bones. In a vertical division of the bones into two groups, account needs to be taken of the fact that the main distribution of bones lies at increasing depth from trench B towards trench D, as was mentioned above in relation to the reindeer bones. If the material from trenches B, H, P, and S is divided such that layer 30-40 is included in the upper part, a hint is seen of the same tendency as in the aforementioned trenches, i.e., an overall decline in sheep in the upper part and a correspond­ ing small rise in goat. Anatomical representation of bones of sheep and goat An anatomical overview of all the bones of sheep and goat is presented in Appendix 3, while the graphical representation in Fig. 12A–C compares the content seen in different parts of the excavation. The survival patterns on dry land (trenches A, F and N) and in the mire are so similar that it must have been the same type of bone waste which ended up in both places. Seen against this background, the refuse in the mire is more likely to be the result of a slow successive accumulation than of larger collective dumps, for example in relation to slaughter. A comparison with the data available on bones of sheep and goat recovered from the building remains at “Gården under Sandet” (Enghoff 2003) reveals TrenchA, F, and N Sheep Goat Layer0­10 22 Layer 10-20 2 2 Layer 20-30 15 (62.5%) 9 (37.5%) Layer 30-40 16 (89%) 2 (11%) Layer 40-50 6 1 Layer 50-60 9 2 Layer60-70 9 2 Layer70­80 2 1 61 (74%) 21 (26%) Upper part (a) 0-10, 10-20, 20-30 19 (59%) 13 (41%) Lower part (a) 30-40, 40-50, 50-60 etc. 42 (84%) 8 (16%) Upper part (b) 0-10, 10-20, 20-30, 30-40 Lower part (b) 40-50, 50-60, 60-70 etc. Trench C, K, and Q Sheep Goat 10 13 (52%) 12 (48%) 15 (45%) 18 (55%) 67 (71%) 28 (29%) 35 (64%) 20 (36%) 3 0 0 0 0 0 134 (63%) 78 (37%) 29 (49%) 30 (51%) 105 (69%) 48 (31%) Trench B, H, P, and S Sheep Goat Trench D Sheep Goat 20 10 9 (60%) 20 (43%) 22 (38%) 33 (50%) 6 (40%) 01 27 (57%) 2 36 (62%) 10 33 (50%) 10 13 (78%) 3 3 13 (57%) 8 (44%) 5 0 0 30 (54%) 17 (57%) 13 (50%) 5 (28%) 1 0 0 0 0 92 (44%) 115 (56%) 31 (48%) 33 (52%) 61 (43%) 82 (57%) 53 (43%) 69 (57%) 39 (46%) 46 (54%) (43%) (56%) 0 26 (46%) 13 (43%) 13 (50%) 21 2018 Journal of the North Atlantic No. 37 G. Nyegaard points of similarity with trenches A, F and N and the mire. Higher proportions of small elements such as carpals, tarsals and phalanges at “Gården under Sandet” are due to the fact that wet sieving was em­ ployed during part of the excavation. Of the long limb bones, there are very few intact examples of, respectively, scapula (4), humerus (1), radius (2), radius-ulna (1), metacarpals (6), and metatarsals (5), in addition to occasional bones of newborn individuals. The summary of the calculated minimum number of individuals (MNI) for upper jaw and lower jaw (molars) and for most of the limb bones augments the anatomical overview and gives an impression of the representation of the various elements (Appendix 4). The calculated MNI values show greater fluctua­ tion from bone to bone than is the case for cattle. This is because the bones of sheep and goats have been exposed to a greater taphonomic load, includ­ Figure 12 A – C. Survival patterns of bones of sheep and goat in trenches A, F, and N (top), in the mire (in the mid­ dle) and in the Norse ditch in trenches Ea and M (bottom). ing the effect of excavation. The fact that it is pos­ sible to distinguish a particularly high number of individuals on the basis of molars from the upper and lower jaw should be seen in the light of the fact that the upper cultural deposits contained a number of teeth originating from disintegrated or damaged jaws. Consequently, from top to bottom, the teeth constituted the following percentage proportions of all remains of sheep and goat: 56 % (layer 0–10), 40 % (layer 10–20), 20 % (layer 20–30), 17 % (layer 30–40), 14 % (layer 40–50), 16 % (layer 50–60), and 13 % (layer 60–70). Age structure of slaughtered sheep and goats An investigation of the age at slaughter of sheep and goats was based on tooth eruption and tooth wear in preserved lower jaw parts with in situ teeth recovered from all the excavations. In advance of this study, remains were identified to species on the basis of the morphological differences described by Payne (1985) for juvenile animals and by Hal- stead and Collins (2002) for adult individuals. As shown by Zeder and Pilaar (2010), identification of mandibles can be very difficult. However, reliability increases when more criteria are used, as was the case with most of the identified 57 mandibles. Fig­ ures 13 and 14 show the age distribution according to mandible wear stage (MWS), as demonstrated by Grant (1982) for sheep and goat, respectively. An investigation of tooth wear and calculation of MWS in a sample of lower jaws from 67 individuals of Gotland sheep of known age in the Zoological Museum’s skeleton collection formed the basis for the approximate age determination of the lower jaw wear stages for sheep given in Figure 13. Half of the total of 30 lower jaws derives from adult animals of more than 2 years of age, where the rear molar, M3, has emerged and begun to wear. The other half derives from lambs slaughtered at different juvenile age-stages; most were slaughtered within the inter­ val 4–12 months. Only a single jaw bone (MWS: 25) is from an animal slaughtered during the second year of its life. This particular individual has an emergent M3 and is therefore close to being fully grown. Four jaw parts (13 %) are from lambs slaughtered within the first 2 months of life. Habermehl (1975) gives more or less the same eruption ages for the 3 molars in both goat and sheep. On the basis of the different lower jaw wear stages in slaughtered goats in Figure 14, some of the same age intervals have been quoted as on Figure 13. At the same time, account has also been taken of the duration of various tooth wear stages in a flock of Turkish Angora goats (Deniz and Payne 1982). Only 22 2018 Journal of the North Atlantic No. 37 G. Nyegaard 7 out of a total of 27 lower jaws of goat are from adult animals (26 %), where the rearmost molar, M3, has emerged and begun to wear. The remaining 20 fall into one of three groups. Nine (33%) are from animals that were slaughtered in their second year of life, while 8 (30 %) are from goat kids slaughtered during their very first months of life. Only 3 (11 %) are from animals slaughtered between about 4 and 7 months of age. Despite the relatively modest number of pre­ served jaws, the investigation indicates such a divergent slaughtering pattern in sheep and goats that this must reflect a significant difference in the economic exploitation of the two species. Judg­ ing from the figures, the farm’s adult population of small ruminants contained more sheep than goats. This is also consistent with the results of the investigations of the post-cranial skeleton outlined above. As this investigation also shows that, on average, more or less equal numbers of lambs and kids were born on the farm, the most likely explanation is that a proportion of the adult sheep flock were kept primarily for wool produc­ tion. Conversely, the high proportion of goat kids slaughtered in the very earliest months of life (30 %) must be seen as showing that the adult goat herd comprised primarily females who were kept for their milk. A left lower jaw of a ca. 1–2 month old goat kid (find number x 1416) has a transverse cut mark from a metal knife on the lower surface of the ramus mandibula, showing that this comes from a slaughtered animal (Fig. 15). The degree to which the sheep’s milk was also exploited is less certain, but the 4 lower jaws from newborn to 2 month­old lambs (13 % of the sheep jaws) suggest that this was actually the case. With respect to both species, a high proportion of young animals were slaugh­ tered, underlining the fact that they both played a role in supplying meat for the farm in addition to their exploitation for wool and milk. Lambs were slaughtered predominantly in their first year of life, whereas a proportion of the young goats were allowed to achieve a greater body weight and were first slaughtered during their second year. These results could be biased somewhat by a group of mandibles with more molars in situ that Figure 15. Left mandibular of a kid with cut marks. Figure 13. Age distribution of slaughtered sheep based on mandibular wear stages (n = 30) (after Grant 1982). Figure 14. Age distribution of slaughtered goats bases on mandibular wear stages (n = 27) (after Grand 1982). 23 2018 Journal of the North Atlantic No. 37 G. Nyegaard could not be identified to species level. MWS could be estimated for 15 mandibles from either sheep or goat slaughtered at these ages: 0–2 months (1 individual), 11–12 months (4 individuals), 12–23 months (3 individuals), ca. 2 years (2 individuals), ca. 3 years (1 individual), 3–4 years (1 individual) and 4 years or older (3 individuals). Furthermore, there are several small mandible fragments with a single tooth in situ. Wear analyses of the loose dp4 and M3 from the lower jaw provide further information on slaughter­ ing age. Figure 16 shows the wear recorded on loose (non-shed) dp4 of sheep and goat, respectively, from all the investigations at the site. Out of 14 dp4 of sheep, 12 show Grant’s wear stages f and g (Grant 1982), which in the above-mentioned collection of Gotland sheep occur in 4–8 month-old animals. The remaining 2 have wear matching animals that are 1–2 years of age. For goat, an almost equal distribution is seen of dp4 from animals slaughtered in, respectively, their first (wear stages f and g) and second (wear stages l and m) year of life. The loose dp4 from the lower jaw show therefore the same ten­ dency as the preserved lower jaws of younger sheep and goats with teeth in situ. A total of 119 loose M3 of sheep/goat were recovered from all the investigations and in most instances these could not be identified to species. It should be emphasised that the large number of teeth includes many stray finds. This contributes to the major difference in the number of loose dp4 and M3, because the milk teeth will more frequently be overlooked. The teeth wear stages according to Grant (1982) are shown in Figure 17. Rather more than half (62) show the very long duration wear stage g which, in the reference collection of Gotland sheep, occurs from around 31⁄2 years and at least until the age of 6, whereas the same wear is first seen in Turkish Angora goats from around the age Figure 16. Tooth wear of loose dp4 of sheep and goat after Grant (1982). of 4 years (Deniz and Payne 1982). A total of 45 teeth showing wear stages b–f are from young adult sheep and goats, of which most must be presumed to have been slaughtered between the ages of ca. 2–4 years, whereas 11 teeth are from slightly younger animals, where M3 had either not emerged or was just beginning to erupt. The relatively large number of teeth from younger adult animals could suggest that the age group 2–4 years is under­represented in the preserved lower jaws that form the basis for Figures 13 and 14. Consistent with the results outlined above, the assemblage includes a greater number of bones of late-juvenile goats than of sheep. This applies for example to metapodials, the distal epiphysis of which is not fused and a small portion of calcaneii of juveniles. Sex determination of sheep and goats Histograms of the bone dimensions for sheep and goat show no clear signs of a bimodality which would allow estimation of the relative frequency of the sexes in the slaughtered adult animals. All that can securely be concluded is that individual bone dimensions which, in the case of both species, deviate in size from the rest must refer to rams and bucks, whereas most of the others probably relate to females. Of the 41 horn cores and core fragments of sheep recovered from the excavations in 1997 and 1998, 7 are from rams. The others originate primarily from female sheep, but a group of 6 cores, midway in size between the typical horn cores of males and females, are presumably from wethers. In form they resemble the cores of female sheep, but internally they have the greater cavity seen in wethers which have been castrated during the first weeks of life (Hatting 1975). On the photograph shown in Figure 18, these horn cores are shown alongside the skull of a 21⁄2 year old Gotland wether from the Zoological Figure 17. Tooth wear of loose M3 of sheep and goat after Grant (1982). 24 2018 Journal of the North Atlantic No. 37 G. Nyegaard Museum’s reference collection (K 438), which was castrated at the age of 3 weeks. From stray finds and other investigations there are a further 18 horn cores of sheep, 6 of which are from rams. The remainder come from female sheep, apart from a single core which, due to its size and internal cavity, could be from a wether. None of the 24 goat core fragments from the investigations in 1997 and 1998 are from bucks. The material recovered from the other investigations at the site includes 2 very robust cores from bucks with their characteristic twisted form, in addition to 10 core fragments from female goats and 1 from a juvenile goat, probably a young male. The sheep and goat horn cores have, in most cas­ es, been cut or broken off close to the base, as is also apparent from the skulls in the assemblage. Cut marks made with metal knives, which demonstrate the use of the actual horns, are seen on cores of both species. Body size of sheep and goat Only a few long limb bones of the two species are intact enough for their greatest length (GL) to be measured. Approximate shoulder height can be calculated from these dimensions using the conver­ sion factors given by, respectively, Teichert (1975) for sheep and Schramm (1967) for goats (Table 9). The estimated shoulder height for sheep, on the basis of this small assemblage of entire bones, var­ ies from ca. 53 to 66 cm. This corresponds to small individuals in other Norse bone assemblages from Greenland, as is apparent from a comparison with the overview given in Enghoff (2003, tables 14–16). The bone dimensions of the sheep metapodials are on a par with the few published measurements from “Gården under Sandet”. Enghoff suggests that the sheep in the Eastern Settlement were slightly larger and slightly less robustly built than those of the Western Settlement, but there is a need for further published sets of measurements from both settle­ Figure 18. Horn cores of wethers. ments before it can be ascertained whether there is actually a difference in the average size of the individuals. The few entire goat bones also originate from very small animals with estimated shoulder heights of ca. 53–60 cm. The other bone elements include examples from very sturdy individuals, demonstrating that the variation in size was con­ siderably greater. For example, a distal joint end of a left tibia (x 1545) has a greatest breadth (Bd) of 30.7 mm. In comparison, the same measurement on one of the largest male goats in the Zoological Mu­ seum’s reference collection (MK 127) is 29.6 mm. The greatest length (GL) of the same individual’s metacarpal is 126 mm. This gives a calculated shoulder height of ca. 72 cm. The horn cores in the assemblage also reveal major gender–related dimorphism in the goats. Some very sturdy goat radii are from late juvenile animals, presumably males. There are also a number of bone fragments of other individuals slaughtered on the threshold to adulthood which cannot be iden­ tified to species as either sheep or goat, but which possibly also originate from young male goats. This applies for example to several parts of the distal diaphysis of the tibia, where the epiphysis is not yet fused. Other Norse assemblages from Greenland also include bones from large robust goats which were slaughtered as late juveniles (Degerbøl 1936). Castrated bucks? The identification of one left metacarpal (x 149) proved problematic. The proportions of the bone indicate goat, even though it is slightly longer than Table 9. Calculation of shoulder height of sheep and goats. Sheep, Ovis aries Radius Metacarpus Metatarsus Goat, Capra hircus Metacarpus Metatarsus Find no. x 1396 x 1568 x 148 x 858 x 1000 x 1528 x 1411 x 1681 x 465 x 148 x 893 x 148 unnumb. x 1399 x 1400 x 1544 x 1612 x 2458 x 1629 x 1634 GL Factor 135.4 4.02 139.6 4.02 149.9 4.02 109.1 4.89 115.7 4.89 126.5 4.89 c. 134.8 4.89 c. 118.5 4.54 c. 126.8 4.54 130.3 4.54 92.9 5.75 c. 93.5 5.75 c. 97.4 5.75 98.2 5.75 104.2 5.75 c. 102.4 5.34 103.9 5.34 105.9 5.34 110.9 5.34 c. 111.9 5.34 Shoulder height (cm) 54.4 56.1 60.3 53.3 56.6 61.9 c. 65.9 c. 53.8 c. 57.6 59.2 53.4 c. 53.8 c. 56.0 56.5 59.9 c. 54.7 55.5 56.6 59.2 c. 59.8 25 2018 Journal of the North Atlantic No. 37 G. Nyegaard that normally seen in goats. It was not possible to find metacarpals of the same proportions among either recent or sub­fossil sheep in the Zoological Museum’s reference collection. There were sheep bones of the same width but then these were lon­ ger. Morphologically, the rear termination of the proximal lateral joint surface is “sheep-like”, but an examination of this character in goats revealed that it can vary considerably and potentially also resemble the bone in question. Several other features of the proximal part also indicate goat. It is striking that the development of the muscle attachments in the upper joint end suggest that the bone come from an adult individual, but at the same time the loose distal epiphysis stands in contradiction to this. This leads thoughts in the direction of a castrated animal, and comparison of the bone with a corresponding example from a modern castrated animal from Lejre (K 134) reveals a similarity. The above-mentioned distal joint end of a left tibia from a very large goat (x 1545) proved only possible to identify after comparison with many individuals of both sheep and goat, as it has char­ acters which point in both directions. On the basis of its great size and its heavily developed muscle attachments, the bone was considered to originate from an older animal. However, on closer exami­ nation it became apparent that, on its reverse, it is still possible to discern the line of fusion between the diaphysis and epiphysis. As the epiphysis had only fused fairly recently, this also leads thoughts in the direction of a castrated animal. Comparison with the above-mentioned modern castrated buck from Lejre revealed so many morphological points of similarity that this bone must be from a cas­ trated buck. Bones showing traces arising from food preparation Almost all the larger cranial fragments of sheep and goat show traces of the skull having been cleaved longitudinally along its median line. This could be a result of the same form of food preparation as still seen today on the Faroe Islands and in Iceland when making the dish svið, also known in Norway as smalahove. In the preparation of this traditional North Atlantic dish, the wool and skin of the sheep’s heads are burnt off over an open flame, after which the heads are cleaved and later boiled for a couple of hours. This form of food preparation has been demonstrated archaeologically in numerous fau­ nal assemblages from the Viking Age and Middle Ages on the Faroe Islands, Iceland and Greenland (McGovern et al 2009:203). On 4 tibias of sheep and goat, a perforation has been made through the bone wall just above the dis­ tal joint end. This must originate from hanging up the back limbs when air drying or maturing the meat. The Domestic Fauna (II): Cattle Anatomical representation of cattle bones Cattle bones make up 15.9 % of all the identi­ fied mammal bones. Appendix 5 shows the ana­ tomical distribution of all 2453 fragments. Smaller elements such as loose teeth, tooth fragments and phalanges make up a very large proportion of the overall total. In the case of many of the other ele­ ments, there is a relatively even distribution, al­ though cranial parts and foot bones, are generally in the majority, whereas vertebrae, ribs and ster­ num in particular are strongly under–represented. An investigation of the horizontal distribution of the various body parts (Fig. 19A–C) shows only minor differences. In trenches A, F and N there is a slightly higher proportion of cranial bones and foot bones than in the mire, whereas in the mire deposits there is in particular a greater content of rib fragments than on dry land. The two uppermost excavation layers in the mire differ from those below by having a greater content of loose teeth. These originate from disintegrated jaw fragments as a consequence of the poorer conditions for pres­ ervation in the upper mire deposits. With respect to anatomical composition, the cattle bones from the ditch fill in trenches Ea and M show greatest similarity with the other parts of the mire. The large limb bones are generally heavily fragmented as a consequence of marrow extraction. Consequently, only 1 entire radius, 4 metacarpals and 2 metatarsals were recovered from the excava­ tions in 1997–1998. In order to give an impression of the representation of the bone elements, an over­ view has been produced of the calculated minimum number of individuals (MNI) on the basis of the up­ per and lower jaws, shoulder blade, pelvis and limb bones (excluding phalanges), which supplements the anatomical overview (Appendix 4). Loose milk teeth and jaw fragments from calves of only a few weeks of age were excluded from the summary used in the comparison with the post-cranial skeleton, as the bones of very young calves are greatly under­ represented in the latter. As can be seen, the robust lower parts of the humerus and tibia are much bet­ ter represented than the upper parts of these bones, whereas in the case of femur only a very few joint ends or parts of these are preserved. Similarly, it is only possible to distinguish a small number of 26 2018 Journal of the North Atlantic No. 37 G. Nyegaard individuals on the basis of the pelvis due to the high degree of fragmentation. If the latter two bones (femur and pelvis) are excluded, the number of individuals identified on the basis of the other elements is very even, ranging from 7 to 13, but often close to 10. In theory, noth­ ing can be concluded from this with respect to the size of the overall herd of cattle from which these bones originate. But the modest number of individu­ als relative to the quantity of bone and, not least, the small gaps in the MNI values for most of the bone elements is consistent with this being a relatively well­represented assemblage of cattle bones arising from slaughtering and kitchen waste on a farm with only a small cattle herd and for which the overall functional period is unlikely to have extended over very many generations. At “Gården under Sandet”, the animal bones were recovered from cultural deposits within the rooms in the centralised farm (Enghoff 2003). The anatomical composition of the cattle bones in this assemblage shows greatest similarities with the Figure 19A – C. Survival patterns of cattle bones in trench­ es A, F and N (top), in the mire (in the middle) and in the Norse ditch in trenches Ea and M (bottom). bones recovered from the settlement area located between the remains of the dwelling and the mire at Ø34; the major deviation is a greater content of vertebrae in the material from “Gården under Sandet”. Age structure of slaughtered cattle Just like the other bones, the upper and lower jaws are strongly fragmented and this imposes some limitations on the investigation of age at slaughter. Including the 2 jaw fragments recovered prior to and after the excavations in 1997–1998, there are a total of 12 upper jaw parts having 1 tooth or more in situ. Six of these are from 0–3 month-old calves with milk teeth showing limited wear, 1 is from a juvenile individual with 3 milk teeth showing full wear, 1 is from an adult with M3 showing wear on the first lobe, and the remaining 4 are from mature individuals with M3 showing heavy wear. The total assemblage of lower jaw parts with in situ teeth from all investigations comprises 19 frag­ ments. Of these, 14 (7 each from the left and right sides, respectively) are from 0–3 month-old calves with milk teeth showing very little wear. The re­ maining 5 (3 from the right and 2 from the left side) are from adult animals. Of these, a small jaw frag­ ment with an in situ premolar (P2) showing no wear is from a young adult individual, whereas the final 4 have tooth wear which indicate that they are from mature animals. The lower jaw fragments have so few molars in situ that it is possible only in a few instances to reach an overall score for lower jaw tooth wear as demon­ strated by Grant (1982). An evaluation of the age at slaughter is therefore based on the dp4 and M3 from the lower jaw, including both loose and in situ teeth from slaughtered animals. Table 10 shows the wear stages that occur as defined by Grant (1982). In theory, a dp4 with one of the later wear stages (g, j, and k) could originate from the same individual as one of the M3 present without wear or in the first wear stage a, cf. table 5 in Legge (1992). In the other cases there is no possibility of an age­related overlap between the teeth in the table. Table 10. Tooth wear in dp4 and M3 of cattle. dp4 8 in situ in jaw fragments, 28 loose teeth abcdefghjklmn Dextra 2732----41--- Sinistra 1 12 1 ­ ­ ­ 1 ­ 2 ­ ­ ­ ­ Total 3 19 4 2 - - 1 - 6 1 - - - M3 All 28 are loose teeth Nowear a b c d e f g h j k k/l/m Dextra 1 1-3-1-2-4-- Sinistra 2 ­111115­211 Total 3 1141217-611 27 2018 Journal of the North Atlantic No. 37 G. Nyegaard Milk molars in cattle are without wear for about the first 4 weeks after birth, as calves do not begin to consume solid food until after 3–4 weeks (Richter 1982). The investigation demonstrates ac­ cordingly that the assemblage’s many dp4 originate from calves that were viable at birth. On the basis of information given by Richter (1982) and Legge (1992), the many teeth showing light wear stage b must be from calves slaughtered when they were 1–3 months of age, whereas the large group of 28 teeth, out of a total of 36 dp4, showing wear stages a–d are from animals slaughtered during the first 6 months of their lives. Of the 28 M3, 9, i.e., about one third, either show no wear or have only the light wear stages a–c. These are from cattle which were either slaughtered just as they were about to reach maturity or slightly later as young adults (ca. 2–4 years). Even though the wear of M3 is related to age, it is not possible to equate precise ages with the individual wear stages as the latter show a degree of variation influenced by different conditions. Sten (2004) has investigated tooth wear in 72 individuals of slaughtered cattle of known age. The long-lasting wear-stage g is typi­ cally found here in animals between 31⁄2 and 6 years of age, while wear stage j is associated with animals in the age group ca. 4–9 years and, especially, 5–8 years. According to Legge (1992), M3 is worn on all 3 lobes (wear stages e–g) in age group 3–6 years, whereas M3 with heavy wear (wear stages j–k) is linked with age group 6–8 years. Although the investigation of age at slaughter is based on the individual examples of dp4 and M3, it is still possible to draw some more general conclu­ sions. A large proportion of the calves born on the farm were slaughtered during the first months of their lives and most commonly around the age of 1–3 months. Another group of animals was slaugh­ tered either as late juveniles or young adults, i.e., slightly before or at the point where the animals had achieved an adult body size. But most of the adult cattle were slaughtered as mature animals, the ma­ jority between the ages of 4 and 8 years. A slaughtering pattern showing such a high proportion of very young calves and an adult popu­ lation dominated by mature animals must reflect a cattle herd with a special focus on milk production. The animals represented in the middle group were slaughtered around the optimal age with respect to meat production. These must be presumed to be the surplus of bulls or bullocks which had reached adult size. These were possibly augmented by a small sur­ plus of heifers that were allowed to reach maturity in order to ensure a certain economic security in production. Further to these, very young calves and the worn-out milk cows and bulls would, of course, have contributed to the meat supply. Cut marks on 2 lower jaws of very young calves show that these came from slaughtered animals. Fig­ ure 20 (find no. x 1193) shows the medial side of the ramus mandibulae, with cut-marks from butchering on the uppermost part. Not unexpectedly, bones of the young calves are strongly under­represented in the remaining bone material. In addition to 103 loose milk teeth from both upper and lower jaw, 128 bone fragments were recorded from calves slaughtered during the first months of their lives, i.e., in total less than 10 % of all the recorded cattle bones. Sex determination Gender determined size differences in the bones of the adult cattle see best expression in the width of the metapodials and, in particular, the distal joint end of the metacarpus (Grigson 1982). However, there is a tendency for the differences to be less in breeds of small size (Uerpmann 1973)—a characteristic of Greenlandic medieval cattle. When we are also deal­ ing with “mixed stock”, i.e., showing a relatively great degree of variation, as described by Degerbøl (1936), this can contribute further to difficulties in determining the gender of the bones, as a significant metric overlap must be expected between the sexes. Measurements of the distal breadth (Bd) of the metacarpus, to the nearest mm, are shown in the histogram in Figure 21. Especially in large cattle breeds, for example those of the Neolithic, this dimension often falls into two size groups, repre­ senting the two sexes (Legge 1981). However, the apparent bimodality seen in the histogram here, as a left and right group, cannot be equated with, respectively, male and female animals. The few bones to the left must, based on an evaluation of Figure 20. Neonatal cattle mandibular with cut marks. 28 2018 Journal of the North Atlantic No. 37 G. Nyegaard the overall assemblage, belong to some particularly small female individuals, whereas the other female animals are to be found in the large group in the right half of the histogram. The scatter diagram, Figure 22, shows the corre­ lation between two relative breadth measurements for 8 entire metacarpals from all the investigations. For the 2 metacarpals measured just after recovery in 1998 and again several years later when they had dried out, two different values are shown. Com­ bined with a visual evaluation, it seems probable that the 4 bones located uppermost to the right are from bulls, whereas the remainder are from cows. The greatest distal breadth of the metacarpus for these 8 bones is marked, respectively, ♂ and ♀ on the histogram, Figure 21. Two of the presumed bulls lie furthest to the right, whereas the 4 cows lie on the left side of the larger cluster. The other 2 presumed bulls have a smaller breadth corresponding to that of the cows, indicating that a considerable degree of overlap can be expected. Even though nothing fur­ ther can be deduced with respect to the quantitative relationship between slaughtered cows and bulls, the data suggest that bulls were well­represented in the adult cattle herd. This is also the impression gained from the other bone elements. Further information on the farm’s cattle herd is provided by 20 horn cores or core fragments re­ covered from all the excavations. A piece of a left frontal bone (square meter 109/103, layer 30–40) has about the lower two thirds of a very robust horn core preserved, which sits on a pronounced “neck” (Fig. 23). The greatest breadth at its base is 5.7 cm and the preserved length is 20 cm. The core follows an upward and outward course relative to the frontal bone and can be assigned to Armitage and Clutton-Brock’s (1976) size category “long horned”. Even though it possesses some of the morphological characters that characterise bull­ ocks in the same authors’ classification system, the oval cross­section and the dense bone tissue of the core suggest that it comes from a bull. A core from the Western Settlement described by Degerbøl (1936:24) shares some similarities. Another fragment of a right frontal bone (x 1684) has a shorter basal part of a robust core correspond­ Figure 22. Scatter diagram showing the correlation be­ tween two relative width measurements of metacarpus of cattle. Two values are shown for two metacarpals mea­ sured both before (open signature) and after drying out. Figure 23. A piece of the left frontal bone of a bull. Figure 21. Histogram of measurements of the distal breadth of metacarpus (Bd) of cattle. 29 2018 Journal of the North Atlantic No. 37 G. Nyegaard ing to the above. Yet another frontal bone fragment from the right side (x 1488) has a large piece of a core attached like the aforementioned, although it is slightly less robustly developed; an open suture between frontal bone and the temporal bone, to­ gether with the core’s slightly porous surface, shows that it comes from an older juvenile or a sub­adult individual. Two entire cores (x 605, x 2460), with a very porous surface, are from juveniles; both are sturdy in form, i.e., very short relative to their basal circumference, and they must be from bull calves. A fragment of frontal bone with the base of a similar core (x 1483) is also assumed to be from a bull calf. Four cores (x 189, x 1485, x 1634 and x 2458) are from cows; these have a greater length relative to their basal circumference, giving them a less robust appearance. With lengths ranging from 14 to 16 cm, they lie on the borderline between Armitage and Clutton-Brock’s (1976) size classes “short horned” and “medium horned”. Two other core fragments are similarly from cows. A couple of robust horn cores in the assemblage show cut marks around their base. Similarly, there are 2 fragments of frontal bone from juvenile or sub­ adult individuals, from which the cores have been cut off. A rather special feature is seen in 4 preserved parts of the outermost part of the horn, which appear whitish and with a clear lamellar structure. One has a circular cross-section (x 2458) and was found in as­ sociation with one of the above­mentioned cow horn Figure 24A. A horn core of a cow with the outer part of the horn preserved. Figure 24B. Three preserved parts of cattle horn. cores, to which it belongs (Fig. 24A). Two of the oth­ ers (x 266 and x 1495) have a flattened cross-section and belong to robust cores which, by all accounts, are from bulls (Fig. 24B). On the basis of the sex determination of some of the bones in the assemblage it can therefore be concluded that the farm had its own cattle breed­ ing, keeping both cows and bulls. With the main emphasis on milk production, a high proportion of cows is to be expected. When the assemblage suggests, nevertheless, that bulls were well repre­ sented, this should be seen in the light of the fact that even one breeding bull in a small cattle herd will figure prominently in the farm’s bone waste. The assemblage also suggests that more bull calves were allowed to live than were needed for breeding purposes. These were then slaughtered as juveniles or sub-adult individuals for their meat. Some of the above­mentioned dp4 showing heavy wear and M3 with no or only slight wear could originate from these young bulls. Shoulder height of cattle With the aid of the metapodials for which the greatest length (GL) can be measured, it is possible to calculate the approximate shoulder height for cattle. In this, use was made of the factors proposed by von den Driesch and Boessneck (1974). The 8 metacarpals in the assemblage were divided up according to the sex determination outlined above (Table 11). It should be remembered that the mea­ surements were carried out on bones, which, after drying, had lost some of their original volume. The greatest length (GL) of 2 metacarpals, find numbers x 1252 and x 1253, was found to be respectively 6.5 mm (3.5 %) and 7.6 mm (4.4 %) less after drying (see above). Table 11. Calculation of shoulder height of cattle. Metacarpus Find no. x 895 x 2460 x 1198 x 162 Mean x 151 x 1253 x 162 x 1252 Mean Metatarsus Find no. x 1410 x 644 x 151 Mean ♀ ♀ ♀ ♀ ♀ ♂ ♂ ♂ ♂ ♂ GL 173.4 176.1 179.4 182.2 166.2 164.9 174.0 181.9 GL 188.4 c. 196.4 c. 204 Factor 6.0 6.0 6.0 6.0 6.3 6.3 6.3 6.3 Factor 5.45 5.45 5.45 Shoulder height (cm) 104.0 105.7 107.6 109.3 106.7 104.7 103.9 109.6 114.6 108.2 Shoulder height (cm) 102.7 c. 107.0 c. 111.2 c. 107.0 30 2018 Journal of the North Atlantic No. 37 G. Nyegaard Based on the length of the metacarpus, the shoul­ der height varies from ca. 104 to 115 cm; the average is 107.4 cm. The 4 presumed bulls extend across the whole range in this very small sample, such that the smallest bull was of about the same height as the smallest cow, while the largest bull was ca. 5.3 cm taller than the largest cow. With respect to size, the cattle from Ø34 are completely on a par with other examples from both the Western and Eastern Settlements, as is apparent from a comparison with the overview provided by Enghoff (2003). Pathology A first neck vertebra, atlas, (x 1640) shows a fracture on its ventral side, Figure 25A–25B. The animal was, however, able to live on with this injury, although without the fracture healing completely. The injury was so incapacitating for the animal that the ventral surface of the vertebra has acquired a “twisted” asymmetrical appearance in that tuber- culum ventrale is displaced from the median line. A probable explanation is that the animal had fallen during grazing in the mountains. Further to this, there are only a small number of cattle bones ex­ hibiting abnormalities or visible traces of illness. These include a fragment of pelvis showing traces of infection around the acetabulum (x 1497) and a Figure 25A. A bovine first neck vertebra with a fracture on the ventral side. Pig A total of 76 bone fragments have been identi­ fied as pig, corresponding to 0.5 % of the identified mammal bones. They show an uneven horizontal distribution, with 51 being found within a ca. 9 m2 area immediately east of the modern drainage ditch in trenches B, H, P and S, but only 3 bones in the trenches on dry land and a total of only 5 in trenches metatarsal with traces proximally of infection across the entire joint surface (x 1410). The Domestic Fauna (III): Horse, Pig, Dog, and Cat As in other Norse assemblages from Greenland there are only few bones of horse, a total 28 frag­ ments, corresponding to only ca. 0.2 % of the identi­ fied mammal bones. Six of the bones are stray finds from alongside the modern drainage ditch. A small concentration of 8 bones is from the ditch fill in trenches Ea and M, while the remaining 14 bones were found scattered in the other trenches. Half the finds comprise loose teeth, the remainder various bone elements (Appendix 6). The only measurable bones are 2 astragalii and 2 calcaneii. These correspond in size to those of mod­ ern Icelandic horses and this is also the case with the horse bones in the other assemblages (Enghoff 2003). There are no clear examples of bones split for marrow extraction, but short cut marks seen under the microscope on fragments of humerus and pelvis, together with tooth marks—perhaps human—on the same two bones indicate that the Christian ban on the eating of horse flesh was not always upheld. Two of the 14 teeth are from juveniles, of which 1 is a fragment of a milk tooth which presumably had been cast. The crown heights of the 4 premolars and molars from, respectively, the lower and up­ per jaw, were measured according to the guidelines described by Levine (1982) in order to provide an approximate age. Four of the teeth proved to be from 5–10 year old animals, 2 are from animals of more than 11 years of age, whereas 1 is from a younger adult individual. Further to these is a single M3 recovered during the small investigation in 1995 which is from a horse of more than 14 year of age. Several small carved wooden toy horses were found in the mire. One of these showed an unmistak­ able resemblance to the Icelandic breed (x 173) (Fig. 26). Horse figures resembling these are also known from a Faroese site dating from the Viking Age or early Middle Ages (Dahl 1979), but have not to date been found elsewhere in Greenland. Horse Figure 25B. A bovine first neck vertebra with a fracture on the ventral side. 31 2018 Journal of the North Atlantic No. 37 G. Nyegaard C, E, K, M, and Q to the north. Vertically, pig bones were found in 6 excavation layers, but just less than half (36 fragments) came from layer 20-30. The 51 fragments from the above–mentioned 9 m2 are dis­ tributed stratigraphically as follows: layer 0–10 (0 fragments), layer 10–20 (3 fragments), layer 20–30 (32 fragments), layer 30–40 (6 fragments) and layer 40–50 (10 fragments). Two bones from the upper­ most excavation layer, layer 0–10, were found in, respectively, squares 103/101 and 107/105. Anatomically, cranial parts (22 fragments) and vertebrae (17 fragments) are the major components. Apart from ribs, most other parts of the body are, however, represented (Appendix 6), so it can be as­ sumed that the pigs were slaughtered on the spot. None of the preserved epiphyses or diaphyses are fused and, based on an evaluation of the surface structure of the other bones, it is striking that almost all are from juvenile animals. The size of the individuals is very small, but only a very few bones could be measured. The only entire bone is a left astragalus, with a greatest length (GL) of 34.6 mm. With the aid of Teichert’s (1969) con­ version factor for this bone (factor 17.9), the shoul­ der height of this animal can be calculated as ca. 62 cm. In comparison, the calculated shoulder height for pigs in Haithabu varies from 57.7 to 88.8 cm, with an average of 70.5 cm (Becker 1980). However, von den Driesch and Boessneck (1974) highlight the uncertainty inherent in the use of small tarsals for the calculation of shoulder height. A right lower jaw with in situ dp2–dp4 and M1, the latter showing only slight enamel wear, is from a ca. 6 month old pig. Further to this, the tooth remains comprise merely a few loose teeth in the form of half a left dp4, a left M1 showing full wear, a piece of a canine and some incisors from juveniles, as well as a single permanent premolar showing light wear. On the basis of this and the other remains, a minimum number of 3 individuals can be arrived at. However, judging from the vertical distribution of the remains, a greater number of individuals is actually represented. Figure 26. A wooden toy horse. L = 7 cm. The assemblage does not suggest that pigs constituted a permanent component of the farm’s livestock, but that now and then over the lifetime of the farm piglets were taken in to be fattened, either from other local farms or through trade with visiting merchant ships from Scandinavia. The lack of remains from adult or mature animals strongly indicates the possibility that no pig breeding at all took place on the farm. Dog The 30 dog bones constitute only a sporadic component in the identified bones of mammals (0.2 %). They were found dispersed in most of the exca­ vation trenches, but with a small concentration of 12 fragments in the ditch fill in trenches Ea and M. Anatomically, the bones represent many different parts of the body (Appendix 6). Eight vertebrae, all with fused epiphyses and recovered from the afore­ mentioned ditch fill, are probably from the same individual. The few preserved jaw parts are too damaged or fragmentary for them to be measured, but the few teeth measurements which were possible (Appendix 10) reveal that the dogs had correspondingly weak sets of teeth as seen in previously investigated as­ semblages (Degerbøl 1936, Enghoff 2003). With regard to the post-cranial skeleton, there are several entire bones. A left humerus (x 1501), with a greatest length (GL) of 203.6 mm, is from a young adult dog of about 1 year of age, as the epiphysis of the upper joint end is only partially fused (Schmid 1972). It is from a large dog with an approximate shoulder height which, according to Harcourt (1974), can be calculated to 67 cm. It therefore belongs to the same size class as the larg­ est dogs recorded from the Western Settlement. As was the case with the latter dogs, this is a slender bone with proportions reminiscent of the long­ limbed greyhound-like type of dog described by Degerbøl (1936). A left radius (x 571) from an adult dog, with a greatest length of 163.2 mm, comes from a slender, medium­sized dog with a calculated shoulder height, according to Harcourt (1974), of ca. 54 cm. The majority of the other bone fragments are from medium-sized to large dogs like the two mentioned above. A left tibia (x 895) from a young adult dog differs from the others in terms of size. The proximal epiphysis of the bone, which is now loose, broke off post-recovery as a consequence of it not being fully fused. This has a greatest length of 94.6 mm which, according to Harcourt (1974), corresponds to a very small dog with a shoulder 32 2018 Journal of the North Atlantic No. 37 G. Nyegaard height of only about 29 cm. The bone has a char­ acteristically curved form with a lateral splaying of the upper half. This particular dog must have been bow­legged, as is the case with many small dog breeds (Fig. 27). Another almost complete tibia (x 1467) is from a young medium-sized dog of about 11⁄2 years of age; the upper epiphysis is missing, whereas the distal epiphysis has recently fused, as the epiphyseal line is visible. The intact bone must have had a greatest length of at least 180 mm, showing that it came from a dog with a minimum shoulder height of 54 cm. A first vertebra, atlas, (x 752) has, due to its size, been identified as either dog or wolf, but examples of large dogs seen in several assemblages speak mostly in favour of the former. The small assemblage of dog bones clearly dem­ onstrates that there were dogs of various sizes and of different breeds. The large dogs were presumably used for hunting or as guard dogs, whereas the small and medium­sized dogs could have been employed by shepherds and goatherds in guarding and manag­ ing the sheep and the goats (Degerbøl 1936). The single bone of a very small dog suggests that not all the animals were necessarily working dogs as this may have been a lap dog. Gnawing marks were seen on an extremely small number of bones in the assemblage, being docu­ mented on only 276 of the 15,455 identified mam­ mal bones (1.8 %); in most instances these were caused by dogs. This low proportion of gnawed bones could be a result of the dogs having spent the summer half of the year on shielings in the farm’s hinterland. It also suggests that when they were on the farm, summer or winter, they did not run about freely, but had a particular place where they would have been tethered. Cat The excavations in 1997 and 1998 produced a to­ tal of 21 bones of domestic cat, corresponding to 0.1 % of the identified bones of mammals. In addition, 5 fragments (all stray finds) were recovered during the evaluation in 1995 and a further fragment was found in 2001 during the teaching excavation, bringing the total of identified bones to 27. Of the 21 bones recovered during the excava­ tions in 1997–1998, 16 were found in the ditch fill in trenches Ea and M, which date from a late phase of the settlement. A femur (x 1701) from here has been radiocarbon dated to the 14th century (see above). The remaining 5 bones are from, respec­ tively, trenches F and N, between the remains of the dwelling and the mire (3 fragments), and trenches O and Q in the mire (2 fragments). The 3 bones from dry land show a great vertical spread: A skull frag­ ment (x 1489) from square 99/101 comes from the uppermost excavation layer (layer 0–10), a fragment of a femur (x 1411) from square 94/101 is from layer 20–30, whereas a lumbar vertebra from square 94/101 is from layer 60–70, i.e., close to the base of the cultural deposits. The cat appears accordingly to have been present both in an early and a late stage in the lifetime of the farm and at least sporadically in the intervening period. The 21 identified fragments represent bones from many different parts of the body (Appendix 6). Further to these are the 6 identified bones from the earlier and later investigations from, respectively, the cranium (3 fragments), lower jaw (1 fragment), femur (1 fragment) and tibia (1 fragment). In the ditch fill in trenches Ea and M an almost complete skull (x 1091) was encountered (Fig. 28). This is plotted in on the section drawing in Figure 10. The left zygomatic arch is missing, together with a canine and the incisors. The open sutures between, respectively, the parietal bone and oc­ cipital bone and parietal bone and temporal bone demonstrate that the skull is from a young adult cat. The ditch fill also yielded a piece of right temporal bone, demonstrating that the 16 fragments recovered here must represent at least 2 individuals. The other bones comprise a right lower jaw half which could belong to the above­mentioned cranium, together Figure 27. Left tibia of a small dog. Figure 28. Parts of two cat skulls. 33 2018 Journal of the North Atlantic No. 37 G. Nyegaard with 3 vertebrae, a right radius, a left femur and a left calcaneum from adult animals—all with fused epiphyses. There are also diaphyses from, respec­ tively, 2 femurs and 2 tibias, presumably from the same juvenile cat, as well as a juvenile metatarsal. Comparisons of the dimensions of the cranium, lower jaws, radius and femur (Appendix 10) show that these first finds of Norse cats in Greenland are, in terms of size, completely on a par with corre­ sponding finds of domestic cat from the Late Viking Age and Middle Ages in Southern Scandinavia as described by Hatting (1990). These animals were rather smaller than modern Danish domestic cats with shoulder heights of around 20 cm (Aaris-Sø­ rensen 1998). On none of the investigated bones was it possible to demonstrate cut marks, as had been the case in the large bone assemblage from a regular cat farm in Odense in Danmark (Hatting 1990). The Wild Fauna Seal Representation of seal species. Collectively, seals constitute the largest animal group in the assemblage, with a total of 6,056 fragments corresponding to 39.2 % of all the identified mammal bones, i.e., a slightly higher proportion than the sum of sheep and goat (36.4 %). All five seal species that are common in South Greenland have been demonstrated in the assemblage. Systematic identification to species has been carried out on elements where identification is possible with greatest certainty, i.e., cranium, lower jaw, humerus, ulna and femur. Use was made of an identification key constructed at the Zoological Museum’s quaternary zoological section, which describes the most characteristic species-specific differences in the above-mentioned elements. The bones of hooded seal are so very different from the other four species that other elements from this spe­ Table 12. Identification to species of seal bones. Common seal Ringed seal cies have occasionally also been identified. A study of the relative economic importance of the five seal species is therefore based on the above­mentioned 5 elements. The figures in Table 12 also give the maxi­ mum figure for the minimum number of individuals (MNI) for each side of each particular bone. Pars tympanica or bulla is one of the most robust elements in the seal skeleton. It is also one of the elements which, due to morphological differences, is best suited to identification to species. This is one reason for there being the greatest number of identi­ fications of this particular bone. Based on the occur­ rence of this element, the five seal species occur in the overall assemblage with the relative proportions shown in Table 13. In taphonomic terms there must be more or less equal chances for the “survival” of this bone ele­ ment. It is therefore presumed that the percentages show a realistic quantitative relationship between the members killed of each seal species. More than 90 % of the bag comprised harp seal and hooded seal, with the former being most dominant, where­ as hunting of the other three species took place only occasionally. Table 14 gives an overview of the horizontal and vertical distribution of the above­mentioned bone elements of harp seal and hooded seal. On the settle­ ment surface (trenches A, F, and N), and particularly in the NE part of the mire (trenches C, K and Q), bones of harp seal are in the majority, whereas the distribution between the two species is more equal in Table 13. Representation of the five seal species present on the basis of the occurrence of pars tympanica/bulla. Common seal, Phoca vitulina Ringed seal, Phoca hispida Harp seal, Phoca groenlandica Bearded seal, Erignathus barbatus Hooded seal, Cystophora cristata Total Harp seal 2032120 2 1 30 1 17 0029120 1 1 23 2 13 012031 0 2 19 3 4 011305 02815 001114 10703 Dextra 2 0 32 1 20 Sinistra Dex. + Sin. % 2 4 3.8 1 1 0.9 30 62 58.5 1 2 1.9 17 37 34.9 55 51 106 100.0 Bearded seal Hooded seal Calvarium Dextra Sinistra Mandibula Dextra Sinistra Humerus Dextra Sinistra Ulna Dextra Sinistra Femur Dextra Sinistra (os temporale) 34 2018 Journal of the North Atlantic No. 37 G. Nyegaard the southern part of the mire (trenches B and G and trenches D, H, I, and P). It is not possible to reach any significant conclusions about the relative abun­ dance of harp seal and hooded seal through time as the figures in Table 14 are too small and too contra­ dictory. In the 3 trench groups to the left of the table, the lowermost occurrences of hooded seal bones are from layers overlying the deepest occurrences of harp seal. It is also apparent from the overview that there is a slight tendency for the relationship between the two species to even out somewhat in the two uppermost excavation layers, layer 0–10 and layer 10–20. In trenches B and G, and trenches D, H, I, and P, respectively, the lowermost three and the uppermost three layers combined show an almost equal distribution of the two species. In trenches Ea and M, which are dominated by the Norse ditch fill dated to a late phase of the settlement, the relative abundance of the two species corresponds more or less to that seen in the assemblage as a whole. Representation of anatomical parts. The overall anatomical summary for all the seal species (Ap­ pendix 7) shows such an even distribution for most of the body parts that it must have been common practice to take whole seal carcasses back to the farm. The same was also true for farms located even further from the coast (Enghoff 2003). There is a significantly higher proportion of seal rib bones (26.9 %) than that seen in the other species—as was also the case at “Gården under Sandet” (Enghoff 2003, table 7). This could be due to the fact that seal ribs are very sturdy and robust. It could also be due to a different way of exploiting seal meat. It is possible, for example, that a seasonal resource such as this could have been used for the production of dried meat. The above overview of the data for the five seal species shows that there has been a greater loss of the hooded seal limb bones than those of the harp seal. This applies in particular to the humerus, of which there are only five. Degerbøl’s (1929) study of seal bones from the excavation at Garðar, the Norse episcopal (bishop) see, in 1926 also revealed a marked “deficit” in the post-cranial bones of hooded seal. Probably this is due to the fact that its bones are less resistant than those of other seal species due to their spongy porous structure, which also makes them very light. However, different butchery prac­ tices could also lie behind. Walrus In all, 48 fragments of walrus bone were identi­ fied, corresponding to 0.3 % of all the identified mammal bones. Apart from a single fragment from trench A, all the bones came from the mire, where they were found distributed throughout all the ex­ cavation layers and in virtually all trenches, with a concentration of 24 fragments being located within a small area in trenches D, H and P. Thirty-nine fragments are from the cranium and only 9 from the rest of the skeleton (Appendix 8). Nine fragments of upper jaw are from the area around the tusks and in 5 cases part of the base of the alveolus is included, where the root of the tusk is attached. One slightly larger jaw fragment (x 1631) shows that the front part of the upper jaw has been cut off just behind the tusks. With respect to the tusks themselves, there are 6 small pieces, of which Table 14. Vertical distribution of identified bones of harp seal and hooded seal. Only fragments belonging to skull, lower jaw, humerus, ulna and femur are included. Trenches A, F and N Trenches C, K and Q Trenches B and G Trenches D, H, I and P Layer Phoca g. Cystophora Phoca g. Cystophora Phoca g. Cystophora Phoca g. Cystophora 0­10 2 2 2 1 0 1 0 1 10­20 41 33 35 40 20-30 8 1 30­40 9 5 40­50 4 1 50-60 00 30 10 83 60-70 00 00 10 00 70­80 10 00 00 00 80­90 00 00 00 00 90­ 0 0 0 0 0 0 0 0 Trench Ea and M Phoca g.: A total of 26 fragments (58%) Cystophora c.: A total of 19 fragments (42 %). 10 3 6 2 2 6 10 2 2 2 11 15 10 3 4 5 10 17 Totals Upper part 0-10, 10-20, 20-30 Lower part 30-40, 40-50, etc. 28 14 (78%) 14 (70%) 10 4 (22%) 6 (30%) 3812 1715 3542 15 (68%) 23 (82%) 7 (32%) 5 (18%) 9 (53%) 8 (53%) 8 (47%) 7 (47%) 6 (46%) 29 (45%) 7 (54%) 35 (55%) 35 2018 Journal of the North Atlantic No. 37 G. Nyegaard 3 are sawn­off pieces of the root end and 1 fragment is the sawn-off tip from a juvenile animal. Further to these are 22 loose premolars and molars from both the upper and lower jaws. An entire left lower jaw, including the front part of the right jaw half, was found in the ditch fill in trench Ea (x 1592). In 1994, 2 cylindrical pieces of bone were recovered from the spoil heaps aris­ ing from the modern drainage ditch. These have a diameter of 5.5 and 5.7 cm, respectively, and have been cut in the plane of symmetry of the front compact part of the lower jaws of 2 adult individu­ als, where the 2 lower jaw halves fuse completely together. These must be presumed to be rough-outs for gaming pieces, as one of the gaming pieces found at the site was made of the same material (Fig. 29A–29B). Gaming pieces were also made from the lateral parts of the lower jaws of walrus (x 1834, x 2065) (Fig. 30). Similar finds have been recovered from other sites (Roussell 1936, Fig. 111). These show that the walrus’ lower jaws were also recovered on hunting trips and taken back to the Eastern Settlement for the production of gam­ ing pieces. As in the case of other assemblages from the Eastern Settlement, there are only a few bones rep­ resenting the post-cranial skeleton, but these do in­ clude a 13 cm long piece of walrus penis bone. Four of the 8 bones are kneecaps. A probable explanation is that these were attached to the sinews of the rear limbs that were taken home from hunting trips. The remaining bones comprise 2 rib fragments, a carpal and a calcaneum. The worked artefacts presumed to be of wal­ rus ivory are not included in the summary. These comprise a side prong of a bird dart from the Early Thule culture (x 172), found at coordinates (x, y) = (103, 101.68), level 241, in trench A (Gulløv 2004: 317), a small carved figure of a polar bear (x 849) from trench D (Fig. 32), 2 carved buckles (x 1321, x 1322) from trench Q (Østergaard 2004, Roesdahl 2015), and several small buttons (x 1376, x 1386, x 1766). One of the buckles seems to have been dis­ carded during manufacture and bears witness to the local production from walrus ivory of slightly more complicated objects. A probable rendition of walrus is seen on a tri­ angular wooden chopping board (x 1386) (Fig. 31). The contours of what appear to be 5 walrus heads sticking out of the water have been carved along one side of it. Whale The assemblage’s total of 80 fragments of whale comprise in particular vertebrae and ribs, in addition Figure 29A. A mandible of walrus together with two rough-outs for gaming pieces. Figure 29B. Part of a gaming piece made from a rough- out of walrus mandible similar those seen on Figure 29 A. Figure 30. Probable rough-outs for gaming pieces made of the lateral part of walrus mandibles. 36 2018 Journal of the North Atlantic No. 37 G. Nyegaard to many small fragments which cannot be identified anatomically. Only 2 fragments could be identified to species and these are both beluga whale. One is a fragment of cranium—a piece of palate bone—from square 105/97 (layer 20–30) in the mire. The other is a tooth from the upper jaw, found in the ditch fill in square 105/109. The same species could be repre­ sented among the total of 13 rib and vertebra frag­ ments which are identified respectively as beluga/ narwhal (6 fragments) and beluga/narwhal/pilot whale (7 fragments). Remains of beluga have been demonstrated at several farms in both the Eastern Settlement (McGovern 1992, McGovern et al. 1993) and the Western Settlement (McGovern 1985, McGovern et al. 1983). The main distribution of the species outside the high­arctic zone is dependent on climatic conditions—the colder the winters, the further south its distribution extends. During a long cold period in the 1850s beluga whales are said to have congregat­ ed in the fjords of South Greenland leading into the Figure 31. A probable rendition of walrus heads on a trian­ gular wooden chopping board. farmsteads of the Eastern Settlement (Vibe 1990). On this basis it seems more likely that the bones in the assemblage originate from animals killed in the region during such relatively cold winters rather than on hunting trips during winter to open­ water areas further to the north in West Greenland. A theoretical possibility of encountering stray animals along all coasts would also always have existed. Pieces of whalebone (baleen) were found in several places during the excavation, and it can be assumed that some of the unidentified whale bones are from baleen whales. During the preliminary investigations in 1994 and 1995, several large fragments of worked vertebrae from large whales were recovered from the spoil heaps produced in creating the modern drainage ditch. In these, the corpus had been hollowed out from one side and the other side shows marks from it having been used as a chopping block or cutting underlay. Two such vertebrae from “Gården under Sandet”, where find circumstances suggest that they were also used as stools, could be identified as Greenland whale/ nordcaper (Enghoff 2003). Polar bear A small assemblage of 19 fragments from polar bear was recovered from both the mire and dry land. It was found distributed through all the excavation layers from layer 0–10 to layer 50–60. Most of the finds are of loose teeth, comprising 5 incisors, 2 canines from the lower jaw, 4 molars also from the lower jaw, a molar from the upper jaw and a further fragment of a canine. The other finds comprise a carpal, a piece of metacarpal, a piece of diaphysis of fibula as well as 3 phalanges, respectively second phalanx (1 fragment) and third phalanx (2 frag­ ments) (Appendix 8). The bones possibly come from animals that were killed locally in the region. The polar bear is a fre­ quent guest in South Greenland and is encountered in particular in spring and summer, during periods when the field ice lies along the outer coast. The metacarpal and the phalanges, which in two instances show cut marks from a metal knife, could have been attached to skins brought home from the hunting grounds. There is also a find of a small plastic figure of a polar bear, carved in tooth (Fig. 32). The head and the body are both individually very true to life, but their relative proportions do not really match up. This could explain why the little figure ended up in the mire. Reindeer The reindeer is the best represented wild terres­ trial mammal in the assemblage, with a total of 975 37 2018 Journal of the North Atlantic No. 37 G. Nyegaard fragments or 6.3 % of the identified mammal bones. Judging from the horizontal (Fig. 11I) and vertical distributions of the bones (see above), most are pre­ sumed to originate from animals killed over a short period within the main settlement period during the 12th and 13th centuries. Prior and subsequent to this period, reindeer hunting only took place to a very modest extent. The overview of the anatomical distribution (Ap­ pendix 8) shows that all parts of the reindeer’s car­ cass were taken back to the farm. The ratio between meaty limb bones (humerus, radius, ulna, femur, and tibia) and lower limb bones (metacarpus, metatarsus and phalanges) is however somewhat greater (1.4) than the corresponding figures for cattle (0.5) and sheep/goat (0.8). This presumably reflects the fact that a proportion of the primary butchering waste was sometimes left at the hunting grounds. It is pos­ sible to establish just as great a minimum number of individuals (13) as for cattle, even though the number of reindeer bones is less than half that of cattle (Appendix 4). There are larger gaps between the minimum number of individuals that can be established on the basis of the various body parts than is the case with the cattle. Again, this could be due to slaughtering waste having been left behind when hunting. A separate overview for trenches H and P (Appendix 4) which, within an area of 6 m2, yielded 409 of the 975 fragment total, reveals a low minimum number of individuals. This is consistent with the assumption that these remains relate to the slaughter and consumption of a small number of animals over a short period of time. The jaw remains are heavily fragmented such that only 17 fragments of upper and lower jaw have one tooth or more in situ. On the basis of tooth erup­ tion data for modern reindeer, a left upper jaw (x 1145) with dp3–dp4 showing wear, M1 in eruption and with slight enamel wear must be from a ca. 2–5 month old calf (referenced in Grønnow et al. 1983). The calf must therefore have been killed either in summer or autumn. A lower jaw with dp4 showing Figure 32. A small plastic figure of a polar bear carved in walrus ivory (L = 3.4 cm). heavy wear and M2 worn on the first lobe is from a young animal killed in its second year, possibly in the summer or autumn. All the other 15 jaw frag­ ments are from adult animals. Most come from 2–5 year old individuals, but 4 with very heavy tooth wear are from even older animals. Juvenile animals are also poorly represented among the loose teeth. The very strong dominance of adult animals is also seen in the other bone remains. These include only 23 fragments of bone showing non-fused epiphyses. A small dataset of bone dimensions can be com­ pared with the data presented by Meldgaard (1986) from the 2 farms in the Western Settlement, V51 and V52a, as well as the Aasivissuit settlement, used by the Thule culture and its descendants. The com­ parison reveals a tendency towards slightly smaller average dimensions in the South Greenland reindeer (Table 15). On the other hand, these differences may be partly due to the demonstrated reduction in volume seen in connection with drying of the bones (see above). South Greenland, south of Ivittuut, has not had a wild population of reindeer since 1800 and there is information suggesting a marked decline in animal numbers prior to the 1750s; this was at about the same time as the major decline in population docu­ mented in West Greenland during the first half of the 18th century (Meldgaard 1986). Reindeer antler has been widely used for mak­ ing tools, including combs and needles/pins. These are not included in the 37 antler fragments shown in Appendix 8, many of which also show traces of working. Arctic hare Eleven bone fragments of arctic hare have been identified. These were recovered both from trenches on dry land (3 fragments) and in the mire (8 frag­ ments). Appendix 8 shows their anatomical distri­ bution. The hunting of arctic hare, possibly with snares, seems to have taken place only sporadically. It is widespread in the area, where studies of its hab­ its around Narsarsuaq have shown that, in summer, it sticks to the various biotopes in the zone between 0–450 m a.s.l. (Thing 1972). Table 15. Comparison of bone dimensions for reindeer. Data from Meldgaard 1986. Humerus: Bd (Mean) Metacarpus: Bd (Mean) Metatarsus: Bd (Mean) Ø 34 45.0 (Σ 6) 39.6 (Σ 5) 37.3 (Σ 6) V51 48.2 (Σ 21) 44.0 (Σ 8) 42.5 (Σ 10) V52a 50.9 (Σ 19) 45.5 (Σ 4) 44.3 (Σ 9) Aasivissuit 47.3 (Σ 13) 43.9 (Σ 17) 43.3 (Σ 22) 38 2018 Journal of the North Atlantic No. 37 G. Nyegaard Arctic fox Most of the 23 fragments of arctic fox are from 2 small concentrations found in, respectively, trench A (10 fragments) and trenches C and Q (7 fragments). Each of these could belong to the same animal, while the remaining 6 fragments were found scattered in the other trenches. Appendix 8 shows their anatomi­ cal distribution. The few bones suggest occasional hunting of arctic fox which could have taken place with snares and stone-built traps. House mouse No bones of mice were recovered, but gnawing marks made by small rodents on a fragment of the diaphysis of a sheep/goat tibia from square 107/98 (layer 40–50) in the mire suggest however that they were present at the farm. This was also the case at other farms in the Western and Eastern Settlements where mouse bones have been found in dwellings and floor layers (Enghoff 2003, McGovern 1992). Birds Of the 313 identified bird bones, 238 come from a modest group of at least 4 species that are all common in the region today (Table 3). The great majority (221 fragments) belong to the auk family, but none have been identified to species as there are no well-preserved remains of beak which enable them to be distinguished reliably. These have been identified as, respectively, common guillemot/Brün­ nich’s guillemot and razorbill/common guillemot/ Brünnich’s guillemot. However, on the basis the present­day distribution of the species, they could all be from Brünnich’s guillemot. At the mouths of the South Greenland fjords in Julianehåb bay there are very large populations of over-wintering Brün­ nich’s guillemot, which come from many different breeding grounds located both to the north and NE. In modern times this species has been one of Green­ land’s most important game animals (Salomonsen 1990). In anatomical terms, the great majority of the bones from the auk family fall into the categories forelimb, shoulder, coracoid bone, collar bone and sternum (Appendix 9). This is also a recurrent fea­ ture of other Norse assemblages (Enghoff 2003). Mallard, white-tailed eagle and rock ptarmigan all breed in the inner South Greenland fjord areas. The latter two species are present all-year-round in the area, whereas the mallard spends the winter in open-water areas along the outer coast. The 2 bones of white­tailed eagle are wing bones found in square 105/99 (layer 20–30). The earliest sub-fossil finds of white­tailed eagle in Greenland are from Norse settlements (Enghoff 2003), but in this respect it is important to add that there are no archaeological sites in South Greenland with well­preserved organ­ ic remains that are earlier than the climatic optimum of the Middle Ages. Apart from 4 fragments of the auk family, all the identified bird bones were recovered from the mire. There are finds of bird bones from all layers from layer 0–10 to layer 50–60, but their occurrence is greatest in layers richest in cultural remains, i.e., layers 20–30, 30–40 and 40–50. It is not possible on the basis of the stratigraphy to conclude any­ thing with respect to how hunting of birds changed through time. Fish The small assemblage of 8 fish bones does not reflect the actual significance of fishing at the site.5 Systematic wet sieving might have given a different result, although the fish bones from the mire would have been further fragmented as a result, due to their poor state of preservation (cf. above). The favour­ able conditions for observing even very small bone fragments during the excavation of the deposits in the mire has however left the excavators with the clear impression that fishing did not play a major role in the economy of the farm. One fishbone is from the area between the mire and the remains of the dwelling (trench F), while the others were found close to each other in the mire in squares 105/101 (layer 50–60) and 106/102 (layer 30–40). Four fragmented vertebrae and a cleithrum are of cod. A fragmented vertebra and a spiny process from a vertebra are of flatfish; the latter is possibly halibut. Further to these is a piece of articulare identified as eelpout. Several species of eelpout are common in the fjords of SW Green­ land, where they are found at very great depths (Muus 1990). Together with the possible bone of halibut, the remains suggest that the inhabitants went fishing with hook and line in the deep water of the nearby fjords. The occurrence of cod along the coast of South Greenland is always assumed to have had a periodic character, whereas in the fjords there were, as now, smaller permanent local popu­ lations (Muus 1990). Mussels In the Norse ditch fill encountered in trenches Ea and M there were a few shells of common mussel. However, pieces of periostracum from disintegrated shells of this species could also be collected in nu­ merous places elsewhere in the mire. The mussels could either have been intentionally gathered or brought to the farm together with seaweed. 39 2018 Journal of the North Atlantic No. 37 G. Nyegaard Summary and Discussion seem to be grounds to conclude that trenches A, F and N, in the activity area between the dwelling and the mire, give an average picture of the species dis­ tribution, whereas the various parts of the mire were more influenced by seasonal and activity-related circumstances. The general picture The assemblage as a whole contains almost equal numbers of bones of domestic (53 %) and wild mam­ mals (47 %), and it reflects an economy that was pre­ dominantly based on sheep, goat and cattle, hunting of harp and hooded seal and—at least in some peri­ ods—hunting of reindeer. The stratigraphic studies reveal, first and foremost, an economy characterised by stability. The differences revealed in the species compositions are relatively modest and an overall picture seems to emerge of a very conservative eco­ nomic system with no dramatic changes through the lifetime of the settlement. The size of the assemblage provides an excellent opportunity to gain an insight into how the domestic animals were exploited. The Norwegian educational text Konungs skuggsjá (The King’s Mirror) from the middle of the 13th century reports that the farms on Greenland produced great quantities of butter and cheese (GHM III 1845:333). Since the very beginning of Norse archaeology, this reference has contributed to a common perception that the exploi­ tation of the secondary products of domestic animals was a driving force in the farms’ economy (Nørlund 1929:139). However, there are to date only very few modern systematic presentations of major bone as­ semblages involving analyses of age at slaughter and a comprehensive dataset of bone dimensions in sup­ port of these conclusions. The assemblage recovered from the excavations of “Gården under Sandet” in the Western Settlement (Enghoff 2003) and analyses of bones from a few other localities in the Western and Eastern Settlements (Mainland and Halstead 2005, McGovern 1992) are some of the most signifi­ cant exceptions to this. Sheep farming and goat herding More than two thirds of the bones of domestic animals in the assemblage are of sheep or goat. Studies of the bones identified to one or other of the species suggest that, on average, more or less equal numbers of kids and lambs were born on the farm, but that the farm’s population of adult animals comprised about twice as many sheep as goats. This is supported by the age determination of the preserved lower jaw fragments. These reveal such a divergent pattern of slaughter that this must At the time of the excavation, this was the first new animal bone assemblage to become available from the Eastern Settlement in almost 40 years. Furthermore, it constitutes the largest faunal assemblage recovered to date from a Norse ar­ chaeological site in Greenland. In the intervening decades, the Western Settlement, located ca. 500 km further to the north, in particular has been sub­ jected to archaeological investigation, resulting in the recovery of several large well­documented assemblages of faunal remains (Enghoff 2003, Mc­ Govern et al. 1983, McGovern et al. 1996). Earlier animal bone assemblages recovered from the East­ ern Settlement are of variable quality (McGovern 1985), and only part of the material from the land­ nam settlement at Narsaq (Ø17a) can be divided up chronologically (McGovern et al. 1993). This new assemblage from the Qorlortoq valley differs from all the previous finds from both of the Norse settle­ ments in that it originates from stratified midden layers within a mire. Representativity The many thousands of bones investigated originate from mire deposits located alongside the entire SE-facing long side of the dwelling in the ruin group, together with an activity area on dry land between the mire and the dwelling. Under any circumstances, this can be presumed to constitute a relatively well­represented sample, providing an excellent insight into the subsistence economy of the farm. Secure evidence on the significance of fishing is not however, for reasons given above, provided by the assemblage. Differences in the species composition are appar­ ent in the various sections of the excavation, particu­ larly in the mire deposits. Especially striking is the concentration of reindeer bones seen in trenches H and P and in parts of trenches B and D. The same 4 square metres which contained the greatest number of reindeer bones in trenches H and P also had the largest contents of goat and hooded seal remains, whereas the greatest content per square metre of cattle remains and bones identified to species as sheep was found in the northern half of the mire. The area with the many reindeer bones also contains the greatest concentration of animal bones as a whole and possibly represents waste thrown directly out from the dwelling—i.e., the composition here could be seasonally influenced and perhaps consists in particular of refuse thrown out during winter, when activities predominantly took place indoors. There 40 2018 Journal of the North Atlantic No. 37 G. Nyegaard reflect a significant difference in the exploitation of the two species. The age profile for the slaughtered goats is con­ sistent with a goat herd which, in the first instance, was exploited for milk. The young goats which sur­ vived their first 3 months or so were often allowed to live on into their second year, whereas lambs that were taken out of production were predominantly slaughtered during their first year. Together with several scattered examples of bones of young goat bucks, slaughtered either as late juveniles or young adults, these young animals must be presumed to have lived out in the mountains all year round, left to their own resources, and representing a meat reserve in the farm’s economy. In addition to the importance of the goats as providers of milk and meat, goat hair was also used in the production of textiles and the skins were important for a number of other purposes (Østergaard 2004:37). Ever since neolithisation there has, in general, been a clear dominance of sheep relative to goat on agricultural settlements in Northern Europe. Norse Greenland therefore provides a rare example of a faunal economy within the Nordic cultural sphere where the two species, in terms of numbers, are of more equal significance (Mainland and Halstead 2005, McGovern 1985). The goat was presum­ ably well adapted to conditions in the two Norse settlements, with their abundant scrub vegetation of willow and birch, and to the local vegetation in general6. The age determination of the relatively small sample of sheep lower jaws indicates a tripartite exploitation involving wool, meat and milk. The even distribution of the preserved lower jaws of, respectively, juveniles (less than 2 years) and adult animals (more than 2 years) is reminiscent of the age distribution for slaughtered animals seen at Neo­ lithic and Bronze Age settlements in South Scandi­ navia (Nyegaard 1985, 1996b), before the advent of the more specialised wool and textile production of Iron Age and Viking Age village communities. However, in contrast to the Bronze Age settlements where the sheep were often slaughtered at an early stage in their adult lives (2–4 years), most of the adult animals reached an age of at least 4–6 years. This conclusion is supported by wear analysis of a large sample of loose rear molars, M3, of which the majority must be presumed to be of sheep. The large and diverse assemblage of archaeo­ logical finds from the mire includes only 2 intact and 2 fragments of soapstone spindle whorls. This is one of the commonest implements found at many other farms which have been investigated, including those in the inland area of the Eastern Settlement known as Vatnahverfi (Vebæk 1992). Together with the age distribution of the sheep, this suggests that there was no particular intensive production of woollen yarn at the farm at Ø34. At most, textile production was only able to meet the occupants’ own requirements. The most widespread trend seen in the variation in the species composition through time is a joint decline in sheep/goat in the uppermost excavation layers. As the stratigraphical studies also show a tendency towards, in the later part of the lifetime of the farm, an evening out of the relationship between sheep and goat, a probable explanation is that the farm’s sheep flock became smaller during this later phase. Østergaard (2004:41) has calculated that a family of 5 needed the wool from 25–30 sheep in order to meet their annual requirement for clothing. On this basis, a decline in the number of inhabitants at a farm will, first and foremost, be expressed in a reduction in the sheep flock. It is therefore possible that there is a connection between the change seen in species composition in the upper layers and a declin­ ing number of inhabitants. An equalling out of the relative numbers of sheep and goat has, incidentally, also been noted at other sites (McGovern et al. 1996). A very rough estimate of the size of the overall sheep flock and goat herd can be arrived at on the basis of a comparison with the calculated minimum number of individuals (MNI) for cattle (Appendix 4). The highest calculated MNI for cattle (13), compared to that for sheep/goat (91), indicates that there were about 7 times as many of the latter as the former. A herd of 5–10 cattle would therefore roughly correspond to a flock/herd of about 35–70 of the small ruminants. The farm’s 2 stone-built animal pens are of the same size and lie next to each other in the southern part of the ruin group (nos. 10 and 11 in Figure 3); these could possibly have been built specifically for sheep and goats. They measure 10 x 5 m and 7 x 7 m, respectively (Guldager et al. 2002: 54), corresponding more or less to a flock/herd of the above-mentioned size. Conversely, the small out­ houses are unlikely to have had the capacity to house such a large number of animals. At several farms in the Western Settlement thick layers of sheep and goat dung have been found in the buildings, bear­ ing witness to the fact that many animals were kept in byres during the winter (Berglund 2000, Roussel 1941). There are also examples from the Eastern Settlement of larger byre buildings having been used for small ruminants, for example the farm associated with Hvalsey fjord’s church, located closer to the outer coast (Roussell 1941). On the basis of the few small ruins of possible sheepcots/goatsheds at Ø34, it must be assumed that many of the animals spent the whole year outdoors. 41 2018 Journal of the North Atlantic No. 37 G. Nyegaard The cattle herd Cattle were of huge significance for the Norse economy. They appear to have been present at all farms, even in the inner regions of the Western Settlement close to the edge of the inland ice cap (Roussell 1941), where the winters were particularly harsh and long and required large stores of winter fodder. In addition to its all-round resource potential in the form of meat, milk, bone, horn, hide etc, the cattle herd was also closely associated with status and perhaps actual cultural identity. Cattle constitute a very stable component in most parts of the excavation. A slight relative increase in the upper layers of the mire is possibly linked to a decline in sheep. A study of the age at slaughter re­ vealed a pattern showing many calves slaughtered at the age of 1–3 months and an adult population com­ prising a majority of mature animals. Even though the post-cranial bones of these young calves are “in deficit”, this pattern can be interpreted as showing that exploitation of the cows’ milk was important. Another group of animals slaughtered either at the late juvenile stage or as young adults, i.e., around the time they became fully grown, shows that meat production was also important. The farm was responsible for its own cattle breeding. By the very nature of things, bulls were just as important as cows in a cattle herd focussed on milk production. A prerequisite for the maintenance of this production was that the dairy cows gave birth to a calf each year. Metric studies suggest that there was a relatively high proportion of male animals; more than was necessary for breeding purposes. These surplus males possibly looked after them­ selves outdoors all year round and were probably members of the group of animals which, according to the tooth remains, were slaughtered as juveniles or young adults. The building presumed to be the farm’s byre lies in continuation of the dwelling in a northerly direction and measures ca. 14 x 5m (no. 3 in Figure 3). The building has thick walls and is divided up internally into two sections which are difficult to interpret. If one of these was used as a cowshed, it could not have accommodated more than 5 to 6 dairy cows. There are now two major, recently analysed faunal assemblages from, respectively, the Western (Enghoff 2003) and the Eastern Settlement (Ø34), which support the reference in Konungs skuggsjá to the exploitation of dairy products in Greenland. Seen in a wider geographical perspective, Norse Greenland was apparently part of a wider North At­ lantic economic tradition with roots extending back into the Iron Age, when the exploitation of milk and dairy products was an important economic strategy (McGovern et al. 2009:190). Other domestic animals The pig is the member of the North European domestic mammal fauna least suited to an existence in the sub-arctic eco-zone. Its bones were found only as a sporadic component in various parts of the excavation and it was not possible to demonstrate the existence of a breeding herd at the site. The ap­ parent almost exclusive presence of bones of young animals is reminiscent of the pattern seen for pigs kept in urban settlements in historical times, where the animals were, for the most, fattened up and eaten when young. From time to time piglets for fattening were presumably acquired by trading either with the few larger farms that had breeding animals, for example the farm at Garðar, the episcopal see (De­ gerbøl 1929), or with visiting Norwegian merchant ships. Examples of pig bones in the upper archaeo­ logical layers suggest that these animals could have been present at the site right up to the end of the farm’s lifetime. As in the case of the other farms that have been investigated, bones of horse are exceptionally few in number. However, this is unlikely to reflect the horse’s true economic significance (McGovern et al.1993). The Viking Age’s many equestrian burials in Scandinavia bear witness to this animal’s special role in Nordic cult and mythology, with traditions extending back to the Early Bronze Age. With the introduction of Christianity, the consumption of horse meat became taboo, and this could explain why horse bones rarely ended up on middens and in floor layers together with the other household waste at Norse farms in Greenland. It is striking, however, that the presence of horse has been dem­ onstrated in many assemblages, and presumably at most farms horses would have been important both as mounts and as draught animals. As is the case now with modern sheep farmers, a horse would have been indispensable during the autumn gathering of the animals. Perhaps it is this vital importance which can be read into the site’s collection of small carved wooden toy horses (Fig. 33). One of these is equipped with a rider, thus documenting the use of the horse as a mount. From ruin group Ø17a in Narsaq (Vebæk 1993) there is also a worked object made of reindeer antler (Fig. 34), which is identical with the cheek pieces of red deer antler for horse bridles found at Bronze Age settlements in Southern Scandinavia and Europe (Nyegaard 1983). Both Ø34 and other ruin groups have finds of presumed whale­ bone sledge runners, which could have been fitted on horse-drawn sledges. 42 2018 Journal of the North Atlantic G. Nyegaard The mortal remains of the farm’s dogs also often ended up elsewhere than on the midden or in the refuse layers. Most of the few dog bones are from medium­sized to large dogs, of which one individual is a candidate for the greyhound-like breed of dog described in several assemblages from the Western Settlement (Degerbøl 1936, Enghoff 2003). One tibia is from a very small bow­legged dog, possibly a lap dog, which has not been seen previously in assemblages from Greenland. Gnawing marks on bones are, incidentally, so rare that it is assumed that the dogs did not run freely around. Furthermore, it is possible that some of the dogs spent the summer at the shielings in the mountains. The bones of domestic cat in the assemblage are the first examples of their kind from the Norse settle­ ments on Greenland. Previously, the cat was only known from a depiction on a presumed wooden arm­ rest from ruin group V51 in Kilaarsarfik (Sandnes) in the Western Settlement. This terminates at the front in a dragon’s head and has 3 cat’s heads carved in relief on its upper surface (Roussell 1936). Even though most of the cat bones are considered to be from the late phase of the farm, their vertical distri­ No. 37 Figure 33. Small carved wooden toy horses. 43 Figure 34. A cheek piece for a horse bridle made of rein­ deer antler from ruin group Ø17a in Narsaq. 2018 Journal of the North Atlantic No. 37 G. Nyegaard bution suggests that cats were also present in earlier times. It is rather strange that cat bones have been found for the first time at a farm of such relatively small size, but this can perhaps be explained by the great size of the bone assemblage. It is possible that cats were also present at other farms and it is not inconceivable that, on more detailed examination of the animal bones in the older assemblages, it will be possible to identify a few bones of domestic cat, for example among the small fragments identified as arctic fox. Frequent finds of house mouse remains at Norse farms (Enghoff 2003, McGovern 1992) show that there would have been a need for the cat’s natu­ ral talents in keeping these under control. Seal hunting A summary based on one anatomical element, pars tympanica or bulla, shows that seal hunting from the farm was predominantly focussed on harp seal and hooded seal. Together, these two species constituted ca. 93 % of the seals killed, and with a proportion of ca. 58 %, the harp seal must be as­ sumed to be the species with the largest number of bones in the overall assemblage. Both are migratory seals which have their main occurrence in Greenland during spring and summer. The harp seal begins to appear along the coast of South Greenland in mid­May, arriving from its breed­ ing and moulting grounds near Newfoundland. Its presence culminates in the region in June–July, when smaller herds are often encountered at the mouths of fjords (Vibe 1990). Harp seals are usually difficult to approach when in herds and we do not know how the Norse hunted them. It has been suggested that use was made of nets placed in strategic locations (e.g., Arneborg 2004:269). According to biologist Aqqalu Rosing-Asvid (pers. comm.) of the Greenland Insti­ tute of Natural Resources, a small population of harp seals (ca. 1000 animals) has bred on the field ice in South Greenland in most years since 2007. If harp seals also bred in the region during the time of the Norse occupation this would have facilitated other hunting strategies, as the female seals tend to stay with their pups to defend them when approached (Rosing-Asvid, pers. comm.). This would also have extended the main hunting season for this species, as in recent years the cubs have been born around 1st April (Rosing-Asvid, pers. comm.). The hooded seal normally lives far out at sea, but comes in to South Greenland at the end of April and in May, on its way to the Denmark Strait where this species gathers on the field ice to moult; the animals return in early August. From a hunter’s point of view, the April–May migration is the most advantageous since it normally coincides with the presence of field ice along the coast of South Green­ land. When the seals are resting on the ice, it may be possible to get very close to them. The Norse hunters would therefore have had the best opportunities to kill hooded seals with their spears, lances and/or bows and arrows at this time of the year7. In summary, it can be stated that the present­day distribution and habits of these two species suggest that most seal hunting took place in the period from April–July in the archipelago along the outer coast, ca. 70–80 km to the SW of the farm. The three other seal species shown to be present are so rare in their occurrence that they must only have been hunted occasionally. The chronological study reveals that seal hunt­ ing was of great economic significance from the very beginning of the farm, and that hooded seals constituted a substantial part of the seal catch throughout almost the entire occupation. In several trenches, the number of seal bones either matches or exceeds that of sheep and goat bones in the lowermost layers. These are often overlain by later layers in which the bones of these small ruminants are in the majority. In the uppermost layers, a clear tendency is seen in several instances towards an in­ creased relative proportion of seal bones. Notably, this development is clearly evident in the 3 trenches on land, whereas the situation in the mire varies somewhat from trench to trench. Entire seals were brought home from hunting trips to the outer coast, as was also the case at other farms (Enghoff 2003, McGovern 1985). The heavy seal corpses could have been transported from the places on the fjords where they were landed by drag­ ging them behind a horse. The seasonal hunting of harp seal and hooded seal must have been carried out jointly by several farms, and it made an important contribution to the household at times of the year when the condition of domestic livestock was at rock bottom. For a com­ munity dependent on an extensively-farmed flock or herd of animals in this part of the sub­arctic cli­ mate zone, where many of the domestic animals are presumed to have lived outdoors all year round, as was the case during the first half century of modern animal husbandry in Greenland from 1924 to ca. 1975, seal hunting could also have constituted an important buffer in the subsistence. Seals constituted a resource that could be exploited even more in the years when people suffered so-called “catastrophic winters” (cf. Madsen 2012). Seen in this light, it seems probable that there was a large fluctuation in the extent of seal hunting from year to year. 44 2018 Journal of the North Atlantic No. 37 G. Nyegaard Much of the seal meat must have been dried for long-term storage. One or more of the ruins of small stone buildings nos. 7–9, Figure 3, on the two mo­ raine ridges just south of the dwelling were probably store houses for provisions such as dried seal meat. Other wild species Bones of walrus are similarly a result of collective hunting and bear witness to the fact that some of men from the farm sometimes took part in organised hunt­ ing trips from the Eastern Settlement, either to more northerly parts of the west coast of Greenland, north of the Western Settlement, or to the southernmost part of the east coast. A new feature demonstrated by finds from the site is that, in addition to upper jaws with in situ tusks, mandibles were also kept to be used in the manufacturing of gaming pieces. The few bones of polar bear which were encoun­ tered could originate from animals killed on the same hunting trips, but could also be from animals which strayed into the local area and were killed there. All in all, reindeer hunting only made a minor contribution to the farm’s subsistence. A greater con­ centration of reindeer bones in one area of the mire could perhaps be seen as reflecting a period during the farm’s lifetime when the opportunities for rein­ deer hunting were better than during the majority of the occupation. It is well known that the reindeer population of West Greenland has, over time, been subject to large fluctuations in numbers (Meldgaard 1986), and it is likely that the same phenomenon also occurred in the population of South Greenland. The concentration of bones could therefore be due to the fact that, for a period, there was a maximum popu­ lation of animals in the areas where people went hunting. These presumably lay in the hinterland, in Johan Dahl Land to the north and NE, or inland to the west or NW. It is also possible that hunting trips were occasionally made to the interior to the north of the Middle Settlement, where there would have been the opportunity to hunt the southern part of the large reindeer population in West Greenland. The catching of arctic hare and arctic fox took place only sporadically. The arctic hare is generally less commonly represented on farms in the Eastern Settlement than in the Western Settlement (Degerbøl 1936). In addition to its meat, there is also an ex­ ample of its fur being used (Østergaard 2004). Bones of arctic fox are generally very few in number on the Norse farms of both settlements (Degerbøl 1936, McGovern 1985). Some catching of guillemot could have taken place in conjunction with spring hunting trips for hooded seal along the outer coast. However, if the flocks of sea birds over-wintering in Julianehåb Bay were as numerous as in recent times, they would have represented such a rich resource that it can be assumed that hunting expeditions took place with the primary objective of bird catching. The method by which the birds were skinned or the way they were processed for consumption must be the reason for the anatomical distribution of the bird bones seen at the site, with a large proportion of, in particular, the elements humerus and pectoral girdle. On the other hand, it is quite remarkable that almost none of the other bones of the bird skeleton have ended up together with the kitchen waste found in the ex­ cavated trenches. Given that today it is possible to catch large num­ bers of arctic char in the late summer, in small pools below the waterfall at the foot of the valley where it meets Tunulliarfik Fjord, it is also very strange that not a single bone of this fish has been demonstrated at the site. Even though sieving was not employed, it is the excavator’s firm conviction that excavation was, in general, so rigorous that a greater number of fish bones would have been recovered insofar as a large amount of fish waste had, in reality, ended up in the mire. Neither were any artefacts found which can be linked to fishing, and this is also true of many other Norse farmsteads (Arneborg 2004:265). The animal bones from Ø34 and other faunal as- semblages from the Eastern Settlement A graphic comparison with the most important of the other published Norse animal bone assemblages from South Greenland is shown in the bar chart in Figure 35. The sites are arranged geographically from left to right, according to their distance from the outer coast. Ø149, on the left, is located on the outer coast, whereas Ø34 to the right is the site lo­ cated furthest inland. The chart reveals the existence of an economic gradient between the outer coast and the hinterland, where the marine element is greatest at the coast and decreases inland. Conversely, domestic livestock increase in relative importance in the opposite di­ rection. The species distribution at Ø34 shows great similarity to that seen at the three Vatnahverfi lo­ calities, Ø71N, Ø71S, and Ø167. The most striking difference is the greater proportion of reindeer bones at Ø34. A corresponding situation is seen at another locality situated on the Narsaq peninsula, Ø17a, and this phenomenon could be due to the two farms, because of their location, having easier access to the good hunting grounds in John Dahl Land to the NE. A geographically-conditioned difference in sub­ sistence between coast and hinterland is also seen in the more inland-oriented Western Settlement, where 45 2018 Journal of the North Atlantic No. 37 G. Nyegaard seal bones at for example Niaqussat (V48) on Amer­ alla Fjord constitute a higher proportion, in percent­ age terms, than at inland farms such as Nipaatsoq (V54) and “Gården under Sandet” (Enghoff 2003, McGovern 1985). As early as 1930, Poul Nørlund, in his publica­ tion of the excavations of the Norse episcopal see Garðar in Igaliku in 1926, broached the idea that seal hunting, with time, came to play an increasing role in the subsistence (Nørlund 1929: 136). It is now a generally accepted fact that, over time, this econom­ ic shift took place at the two Norse settlements and that, notably in the case of the Eastern Settlement, seasonal hunting of the two migratory species, harp seal and hooded seal, became intensified in the Late Middle Ages (Dugmore et al. 2012). According to Arneborg et al. (2008), the diet during the first centu­ ries of the Norse settlement was primarily terrestrial, but from the middle of the 13th century onwards, radical changes took place such that, in the final centuries in the settlement history, the diet became predominantly marine. This conclusion is based on carbon (C) and nitrogen (N) isotope records from a large body of human bone material (Arneborg et al. 2012) and is supported by a few stratigraphically differentiated animal bone assemblages, in which the relative proportions of seal bones increase in the later phases (Enghoff 2003, McGovern 1985). The human skeletons from the first two centu­ ries of Norse settlement in the Eastern Settlement, included in the isotope studies, originate almost exclusively from the central settlement area in the hinterland, whereas the majority of the skeletons from the final two centuries of occupation are from Figure 35. A graphic comparison with other Norse animal bone assemblages from South Greenland based on data from Vebæk 1991 (Ø149), McGovern et al. 1993 (Ø17a) and McGovern 1992 (Ø71N, Ø71S, Ø167). from the coastal zone, i.e., the southern part of the Eastern Settlement (Arneborg et al. 2012). With reference to Figure 35, there is therefore reason to believe that there were, to a great extent, also economic differences in adaptation between, respec­ tively, the outer coast and the hinterland—over and above those chronologically­determined differences in the subsistence base—which see expression in the results of the isotope studies. Further isotope analyses of human skeletons from, respectively, the later occupation in the hinterland and from the early occupation at the outer coast are therefore high on the wish list with respect to future studies. The assemblage from Ø34 shows, through mate­ rial from almost every trench, that seal hunting was important from the very beginning of the farm’s life­ time. The stratigraphic studies of animal bones from trenches A, F and N, located between the mire and the dwelling, also show that seal hunting achieved greater economic significance in the farm’s late phases. Con­ versely, the picture in the mire trenches is rather more varied, partly as a consequence of differences in the horizontal distribution of the species. With the caveat that the final century of the history of the Eastern Settlement is probably not represented at the site, there is however nothing in the faunal assemblage in­ vestigated from Ø34 in Qorlortup Itinnera to suggest that radical economic changes took place during the course of the farm’s lifetime. Rather the opposite. Acknowledgements I would like to thank “Farumgaard Fonden”, SILA - The Greenland Research Centre at the National Museum of Denmark, and what was then the Greenland Home Rule (Pujlen for Forskningsfremme) for grants which made this study possible, and also “Letterstedtska Föreningen” for a financial contribution to the translation into English. Furthers thanks go to my former employers at the Qaqortoq Municipality in South Greenland (now a part of Kommune Kujalleq) and the Greenland National Museum and Archives. I addition, I would like to thank many individuals, including: Knud Rosenlund, for his great help with plans, sections and other of the illustrations, and for his persistent assistance during the study periods at the Natural His­ tory Museum of Denmark (The Zoological Museum), Copenhagen. Jeppe Møhl, the Zoological Museum, for his initial iden­ tification to species of the two rough-outs for gaming pieces in Figure 29A. Geert Brovad, the Zoological Museum, who has taken the photographs in Figure 26 and 28. Kasper Lykke Hansen, the Zoological Museum, who has taken the photographs in Figures 15, 18, 20, 23, 24 A-B, 25 A-B, and 27. 46 2018 Journal of the North Atlantic No. 37 G. Nyegaard Sheep farmers Tippu and Karl Kleist, Qorlortup Itinnera, for their hospitality and help during the many weeks of archaeological excavation in their beautiful valley. Jette Arneborg, The National Museum of Denmark, for long and good collaboration during the field work at Ø34 and at other Norse localities in Greenland. Charlie Christensen, the National Museum of Denmark, for his reading of the chapter about the geology of the mire, and for his comments. David Earle Robinson, for the translation into English. And finally, everyone who participated in the investiga­ tions of the mire at ruin group Ø34. Literature Cited Aaris-Sørensen, K. 1998. Danmarks forhistoriske dyrever­ den. 3. oplag. Gyldendal, København, Denmark. 232 pp. Armitage, P. L., and J. Clutton-Brock. 1976. A system for classification and description of the horn cores of cattle from archaeological sites. Journal of Archaeolo­ gical Science 3(4):329–348. Arneborg, J. 2004. Det europæiske landnam. Nordboerne i Grønland. Pp. 221–278, In H.C. Gulløv (Ed.). Grønlands Forhistorie. Gyldendal. København, Denmark. 434 pp. Arneborg, J, J. Heinemeier, and N. Lynnerup. 2008. “Husk at folk lever af flere ting end bare af brød.” Om de norrøne grønlænderes kost. Nationalmuseets Arbejds­ mark 2008:149–160. Arneborg, J., N. Lynnerup, and J. Heinemeier. 2012. Hu­ man diet and subsistence. Patterns in Norse Greenland AD c.980–AD c.1450: Archaeological interpreta­ tions. Journal of the North Atlantic, Special Volume 3:119–133. Becker, C. 1980. Untersuchungen an Skeletresten von Haus- und Wildschweinen aus Haithabu. Berichte über die Ausgrabungen in Haithabu, Bericht 15. Wachholtz Verlag, Neumünster, Germany. 96 pp. Berglund, J. 2000. The farm beneath the sand. Pp. 295– 303, In W.W. Fitzhugh and E.I. Ward (Eds.). Vikings. The North Atlantic Saga. Smithsonian Institution Press. 432 pp. Boessneck, J., H.H. Müller, and M. Teichert. 1964. Osteo­ logische Unterscheidungsmerkmale zwischen Schafe (Ovis aries Linné) und Ziege (Capra hircus Linné). Kühn-Archiv 78. Heft 1/2. Akademie Verlag. Berlin, Germany. 129 pp. Calib 2012. http://intcal.qub.ac.uk/calib/. Accessed July 2012. Christensen, C. 2000. Preliminary report on the geology and stratigraphy of the mire at ruin group Ø34. Sent to the author, who was at the time curator at Qaqortoq Museum, South Greenland. Dahl, S. 1979. Forn barnaleiku í Føroyum. Mondul 1979:3, 3–13. Tórshavn, Faroe Islands. Degerbøl, M. 1929. Animal bones from the Norse ruins at Gardar, Greenland. Meddelelser om Grønland 76:183–192. Degerbøl, M. 1936. Animal remains from the West Settle­ ment in Greenland with special reference to livestock. Meddelelser om Grønland 88(3):1–54. Deniz, E. and S. Payne. 1982. Eruption and wear in the mandibula dentition as a guide to ageing Turkish Angora goats. Pp. 155–205, In B. Wilson, C. Grigson, and S. Payne (Eds.). Ageing and Sexing Animal Bones from Archaeological Sites. BAR British Series 109. 268 pp. Dugmore, A.J., T.H. McGovern, O. Vésteinsson, J. Arneborg, R. Streeter, and C. Keller. 2012. Cultural adaptation, compounding vulnerabilities and conjunc­ tures in Norse Greenland. Proceedings of the National Academy of Science of the United States of America 109:3658–3663. Enghoff, I.B. 2003. Hunting, fishing and animal hus­ bandry at the Farm Beneath The Sand, Western Greenland. Meddelelser om Grønland. Man & Soci­ ety 28. 104 pp. Gabler, K.-O. 1985. Osteologische Unterscheidungsmerk­ male am postkranialen Skelett zwischen Mähnen­ springer (Ammotragus lervia), Hausschaf (Ovis aries) und Hausziege (Capra hircus). Inaugural-Dissertation zur Erlangung der tiermedizinischen Doktorwürde der Tierärtzlichen Fakultät der Ludvig-Maximilians-Uni­ versität München, Germany. 127 pp. GHM III. 1845. Grønlands Historiske Mindemærker (GHM) 1838–1845. I–III. Det Kongelige Nordiske Oldskriftselskab. Det Brünnichske Bogtrykkeri, Kjø­ benhavn, Denmark. Facsimile 1976. Rosenkilde og Bagger, København, Denmark. 950 pp. Grant, A. 1982. The use of tooth wear as a guide to the age of domestic ungulates. Pp. 91–108, In B. Wilson, C. Grigson, and S. Payne (Eds.) 1982. Ageing and Sexing Animal Bones from Archaeological Sites. BAR British Series 109. 268 pp. Grigson, C. 1982. Sex and age determination of some bones of domestic cattle: A review of the literature. Pp. 7–23, In B. Wilson, C. Grigson, and S. Payne (Eds.). Ageing and Sexing Animal Bones from Archaeologi­ cal Sites. BAR British Series 109. 268 pp. Grønnow, B., M. Meldgaard, and J.B. Nielsen. 1983. Aa­ sivissuit: The Great Summer Camp. Archaeological, ethnographical and zoo­archaeological studies of a caribou-hunting site in West Greenland. Meddelelser om Grønland. Man & Society 5. 96 pp. Guldager, O., S. Stummann Hansen, and S. Gleie. 2002. Medieval Farmsteads in Greenland. The Brattahlid region, 1999–2000. Danish Polar Center. Copenhagen, Denmark. 142 pp. Gulløv, H.C. 2004. Nunarput. Vort land. Thulekulturen. Pp. 281-343, In H.C. Gulløv (Ed.). Grønlands Forhi­ storie. Gyldendal. København, Denmark. Pp. 434. Habermehl, K.-H. 1975. Die Altersbestimmung bei Haus- und Labortieren. 2. Aufl. Verlag Paul Parey. Hamburg- Berlin, Germany. 216 pp. Halstead, P. and P. Collins. 2002. Sorting the sheep and the goats: Morphological distinctions between the man­ dibles and mandibular teeth of adult Ovis and Capra. Journal of Archaeological Science 29:545–553. 47 2018 Journal of the North Atlantic No. 37 G. Nyegaard Harcourt, R.A. 1974. The dog in prehistoric and early historic Britain. Journal of Archaeological Science 1(2):151–175. Hatting, T. 1975. The influence of castration on sheep horns. Pp. 345–351, In A.T. Clason (Ed.). Archaeolog­ ical Studies. Amsterdam, Netherlands. Hatting, T. 1990. Cats from Viking Age Odense. Journal of Danish Archaeology 9:179–193. Høegh, E. 1988. Erling Høeghs håndskrevne erindringer om sin opvækst i Qaqortoq. /Handwritten memoires by Erling Høegh about his childhood in Qaqortoq. A copy kept at the Qaqortoq Museum, Greenland. Krogh, K. 1982. Erik den Rødes Grønland. Nationalmu­ seets Forlag. København, Denmark. 266 pp. Legge, A. J. 1981. Aspects of Cattle Husbandry. Pp. 169– 181, In R. Mercer (Ed.). Farming Practice in British Prehistory. University Press. Edinburgh, UK. Legge, A. J. 1992. Excavations at Grimes Graves, Nor­ folk 1972–1976. Fascicule 4. Animals, environment and the Bronze Age economy. British Museum Press. London, UK. 87 pp. Levine, M. A. 1982. The use of crown height measure­ ments and eruption-wear sequences to age horse teeth. Pp. 223–250, In B. Wilson, C. Grigson, S. Payne (Eds.). Ageing and Sexing Animal Bones from Ar­ chaeological Sites. BAR British Series 109. 268 pp. Madsen, C. K. 2012. Pastures Found: Farming in Green­ land (re)introduced. Pp. 142–166, In H.C. Gulløv, P.A. Toft, and C.P. Hansgaard (Eds.). Challenges and solutions. Northern Worlds. Report from workshop 2 at the National Museum, 1 November 2011. Copenha­ gen, Denmark. 272 pp. Mainland, I. and P. Halstead (2005). The Economics of sheep and goat husbandry in Norse Greenland. Arctic Anthropology 42(1):103–120. McGovern, T. H. 1985. Contributions to the paleoeco­ nomy of Norse Greenland. Acta Archaeologica 54(1983):73–122. McGovern, T.H. 1992. The Zooarchaeology of the Vatnahverfi. Pp. 93–109, In C.L. Vebæk 1992. Vatnahverfi. An inland district of the Eastern Settle­ ment in Greenland. Meddelelser om Grønland. Man & Society 17. 132pp. McGovern, T.H., P.C. Buckland, D. Savory, G. Sveinbja­ rnardóttir, C. Andreasen, and P. Skidmore. 1983. A study of the faunal and floral remains from two Norse farms in the Western Settlement, Greenland. Arctic Anthropology 20:93–120. McGovern, T.H., G. F. Bigelow, T. Amorosi, J. Woollet, and S. Perdikaris. 1993. The zooarchaeology of Ø 17a. Pp. 58–72, In C.L. Vebæk 1993. Narsaq: A Norse landnáma farm. Meddelelser om Grønland. Man & Society 18. McGovern, T.H., T. Amorosi, S. Perdikaris, and J. Woollett. 1996. Vertebrate zooarchaeology of Sandnes V51: Economic change at a chieftain’s farm in West Greenland. Arctic Anthropology 33(2):94–121. McGovern, T.H., S. Perdikaris, I. Mainland, P. Ascough, V. Ewens, A. Einarsson, J. Sidell, G. Hambrecht, and R. Harrison. 2009. Chapter 4: The archaeofauna. Pp. 168–252, In G. Lucas (Ed.). Hofstadir: Excavations of a Viking Age Feasting Hall in North-Eastern Iceland. Institute of Archaeology, Reykjavik. Monograph No. 1, Reykjavik, Iceland. Meldgaard, M. 1986. The Greenland caribou: Zoogeogra­ phy, taxonomy, and population dynamics. Meddelelser om Grønland, Bioscience Vol. 20. 88 pp. Muus, B.1990. Fisk. Pp. 23–158, In B. Muus, F. Salomon­ sen, and C. Vibe. 1990. Grønlands Fauna. Fisk, fugle, pattedyr. Gyldendal. København, Denmark. Nelson, E. 2008. A survey of AMS-datings of bones of terrestrial animals from the excavations at ruin group Ø34. Received in an e-mail to the author at the Green­ land National Museum & Archives on March 18, 2008 from Professor Erle Nelson, Simon Fraser University, Vancouver, BC, Canada. Nyegaard, G. 1983. Dyreknogler fra bronzealderens bopladser i Sydskandinavien. Et studie over fau­ naøkonomi samt bearbejdede genstande af ben og tak, med udgangspunkt i to nye fund. Upubliceret specia­ leafhandling, Københavns Universitet. Unpublished master’s thesis at Institute of Archaeology, University of Copenhagen, Denmark. Nyegaard, G. 1985. Faunalevn fra yngre stenalder på øerne syd for Fyn. Pp. 426–457, In J. Skaarup 1985. Yngre stenalder på øerne syd for Fyn. Meddelelser fra Langelands Museum. Rudkøbing. 491 pp. Nyegaard, G. 1996a. 61V3-III-525 (Ø34) i Qorlortup Itinnera. Arkæologisk undersøgelse fra d. 19.- 24.8. 1995 af norrønt udsmidslag. Unpublished report. Februar 1996. Qaqortup Katersugaasivia/Qaqortoq Museum, Greenland. Nyegaard, G. 1996b. Faunalevn fra bronzealder. En zooarkæologisk undersøgelse af sydskandinaviske bopladsfund. Unpublished PhD thesis. Institute of Archaeology, University of Copenhagen, Denmark. Nørlund, P. 1929. Norse ruins at Gardar. The episcopal seat of mediaeval Greenland. Meddelelser om Grøn­ land 76:7–170. Østergård, E. 2004. Woven into the Earth. Textiles from Norse Greenland. Aarhus University Press, Aarhus, Denmark. Pp. 296. Payne, S. 1969. A metrical distinction between sheep and goat metacarpals. Pp. 295–305, In P.J. Ucko, and G.W. Dimbleby (Eds.). 1969. The domestication and ex­ ploitation of plants and animals. Duckworth, London, UK. Payne, S. 1985. Morphological distinctions between the mandibular teeth of young sheep, Ovis, and goats, Capra. Journal of Archaeological Science 12(2):139– 147. Prummel, W., and H.J. Frisch. 1986. A guide for the distinction of species, sex and body side in bones of sheep and goat. Journal of Archaeological Science 13:567–577. Richter, J. 1982. Faunal remains from Ulkestrup Lyng Øst. A hunter ́s dwelling place. Pp. 141–177, In K. Ander­ sen, K. Jørgensen, and J. Richter. Maglemosehytterne ved Ulkestrup Lyng. Nordiske Fortidsminder, Nr. 7. København, Denmark. 48 2018 Journal of the North Atlantic No. 37 G. Nyegaard Roesdahl, E. 2015. Fine belt-buckles of walrus ivory— also made in Greenland. Pp. 267–273, In I. Bang, J. Larsen and S.S. Mygland (Eds.). Nordic Middle Ages – Artefacts, landscapes and society. Essays in honour of Ingvild Øye on her 70th birthday. UBAS. University of Bergen, Archaeological Series 8. Roussell, Aa. 1936. Sandnes and the neighbouring farms. Meddelelser om Grønland 88(2):3–219. Roussell, Aa.1941. Farms and churches in the Mediaeval Norse settlements of Greenland. Meddelelser om Grønland 89(1):3–342. Salomonsen, F. 1990. Fugle. Pp. 159–361, In B. Muus, F. Salomonsen, and C. Vibe. Grønlands Fauna. Fisk, fugle, pattedyr. Gyldendal. København, Denmark. Schofield, J.E., K. J. Edwards, and C. Christensen. 2008. Environmental impacts of Norse landnám in the Qor­ lortoq valley, Eastern Settlement, Greenland. Journal of Archaeological Science 35:1643–1657. Schmid, E. 1972. Atlas of animal bones. Elsevier Publish­ ing Co., Amsterdam, Netherlands. 166 pp. Schramm, Z. 1967. Long bones and height in withers in goat. Roczniki Wyzszei Szkoly Rolniczej w Poznaniu 36:89–105. Sten, S. 2004. Bovine teeth in age assessment, from Medi­ eval cattle to Belgian Blue. Methodology, possibilities and limitations. Theses and Papers in Osteoarchaeolo­ gy 1. Stockholm University, Sweden. 178 pp. Stuiver, M. and P.J. Reimer. 1993. Extended 14C database and revised CALIB radiocarbon calibration program. Radiocarbon 35:215–230. Takahashi, C.M., Nelson, D.E. 1999. Radiocarbon dating and stable isotope analysis of archaeological faunal remains from Ø34, Greenland. Report #99-3. Archae­ ometry Laboratoy. Simon Fraser University. Canada. Internal preliminary report sent to the author at Qaqor­ toq Museum, Greenland. 16 pp. Teichert, M. 1969. Osteometrische Untersuchungen zur Berechnung der Widerristhöhe bei vor- und frühgeschichtlichen Schweinen. Kühn-Archiv Band 83:237–292. Teichert, M. 1975. Osteometrische Untersuchungen zur Berechnung der Widerristhöhe bei Schafen. Pp. 51–69, In A.T. Clason (Ed.). Archaeological Studies. Amsterdam, Netherlands. Thing, H. 1972. Iagttagelser af snehare (Lepus articus por­ sildi) ved Narssarsuaq 26.6.–26.8.1971. Tidsskriftet Grønland 1972:265–274. Uerpmann, H.P. 1973. Animal bone finds and economic archaeology: A critical study of “osteo-archaeologi­ cal” method. World Archaeology 4:307–332. Vebæk, C.L. 1991. The church topography of the Eastern Settlement and the excavation of the Benedictine con­ vent in Uunartoq Fjord. Meddelelser om Grønland, Man & Society 14:1–81. Vebæk, C.L. 1992. Vatnahverfi. An inland district of the Eastern Settlement in Greenland. Meddelser om Grøn­ land, Man & Society 17:1–132. Vebæk, C.L. 1993. Narsaq: A Norse landnáma farm. Med­ delelser om Grønland, Man & Society 18:1–85. Vibe, C. 1990. Pattedyr. Pp. 363–459, In B. Muus, F. Sa­ lomonsen, and C. Vibe. Grønlands fauna. Fisk, fugle, pattedyr. Gyldendal. København, Denmark. von den Driesch, A. 1976. A guide to the measurement of animal bones from archaeological sites. Peabody Museum Bulletin 1. 136 pp. von den Driesch, A., and J. Boessneck. 1974. Kritische Anmerkungen zur Widerristhöhenberechnung aus Längenmassen vor- und frühgeschichtlicher Tier­ knochen. Säugetierkundliche Mitteilungen 22. Heft 4:325–348. München, Germany. Zeder, M.A., and S.E. Pilaar 2010. Assessing the reliabil­ ity of criteria used to identify mandibles and mandib­ ular teeth in sheep, Ovis, and goats, Capra. Journal of Archaeological Science 37:225–242. 49 2018 Journal of the North Atlantic No. 37 G. Nyegaard Endnotes for metapodials of cloven­hooved animals they were supplemented with numerical codes for, respectively, the medial and lateral halves. 4 It should be noted that the lower frequencies in the south­ ern part of trench S are due to the fact that the excavation here had only reached as far as the base of layer 30­40 when it ended in 1998. 5 The identification of fish bones was carried out by Knud Rosenlund, the Zoological Museum in Copenhagen, who also took part in the excavations in 1997 and 1998. 6 Later in history, during European colonisation, goats were again kept in many places in South Greenland. Intriguingly, the Greenlandic master smith in Qaqortoq, John Høegh, Ujuukulooq, kept a large number of goats during the 1920s and 30s and supplied milk to the town’s hospital among other places. During the day, the animals were taken out into the mountains by Høegh’s children, but were otherwise housed for most of the year in a goat house next to the family’s home in the middle of town (Høegh 1988, 170). 7 According to biologist Aqqalu Rosing-Asvid of the Greenland Institute of Natural Resources, it was possible during a recent tagging study of hooded seals to surprise and catch 30 individuals by throwing a net over them while they were sleeping on the ice (personal communication). 1 The rescue excavation in the mire at ruin group Ø34 was carried out by Qaqortoq Museum from 21st July to15th August 1997 and from 19th July to 22nd August 1998 and led by this museum’s current curator, Georg Nyegaard, with assistance from the National Museum of Denmark and the Zoological Museum in Copenhagen, together with other archaeologists and environmental archaeologists from Iceland, Great Britain and Canada. The final week of the excavation in 1998 took the form of a teaching excavation for two classes from Sydgrønlands Gymna- siale Skole (South Greenland High School) in Qaqortoq, whereby teams of pupils took it in turns to participate in the excavation of trenches P, Q, and S (Fig. 7). In 2001, the excavation was resumed between the 18th and 25th August in the form of a week-long teaching excavation for two classes from Sydgrønlands Gymnasiale Skole in Qaqortoq. This was carried out in collaboration between Qaqortoq Museum, SILA (Center for Greenland Research) at the National Museum of Denmark in Copenhagen and the Greenland National Museum and Archives. A fur­ ther 16 m2 was excavated on the periphery of the earlier trenches in the mire and according to the same methods. A large quantity of finds was again recovered, including ani­ mal bones, but these were not included in the systematic investigations. Similarly, these excavation trenches are not shown on the site plan. 2 The geological studies were carried out by geologist Charlie Christensen of the National Museum of Denmark. The description of the mire stratigraphy is based on his preliminary report sent to the author in 2000 (Christensen 2000). 3 For example, in the case of the humerus, the following numerical codes were employed: Unspecified fragment 0 Entire bone 1 Proximal part 2 Proximal part + 1⁄2 3 Proximal part + 3⁄4 4 Entire bone excl. distal epiphysis 5 Distal part 6 Distal part + 1⁄2 7 Distal part + 3⁄4 8 Entire bone excl. proximal epiphysis 9 Mid­piece 10 Proximal epiphysis (juvenile) 11 Distal epiphysis (juvenile) 12 Entire diaphysis (juvenile) 13 Proximal diaphysis (juvenile) 14 Distal diaphysis (juvenile) 15 The minimum number of individuals (MNI), as calculated on the basis of the distal part of left humerus of a species, consist of the sum of elements with the numerical codes 1, 6, 7, 8, and 9 plus either the sum of elements with the codes 5, 13 and 15, or the number of elements with code 12; only the largest of the last two values are selected to avoid an individual being counted more than once. The codes for radius, femur and tibia were the same, whereas 50 2018 Journal of the North Atlantic G. Nyegaard Appendix 1 A. Distribution of animal bones in excavated trenches and layers. No. 37 Totals for A, F, N, 103/101 NISP % 31 11.7 55 20.7 176 66.2 4 1.5 266 100.1 28 11.0 70 27.6 151 59.4 5 2.0 254 100.0 96 18.8 233 45.6 177 34.6 5 1.0 511 100.0 2 72 15.3 206 43.7 187 39.7 6 1.3 471 100.0 30 12.8 98 41.7 101 43.0 6 2.6 235 100.1 16 11.7 63 46.0 5741.6 1 0.7 137 100.0 11 15.5 31 43.7 29 40.8 0 0.0 71 100.0 Settlement area Layer 0–10 Layer 10–20 Layer 20–30 Layer 10–30 Layer 30–40 Layer 40–50 Layer 50–60 Layer 60–70 Layer 70–90 Trench NISP 0 7 15 0 22 8 9 47 2 66 52 99 77 2 230 42 114 83 4 243 3 15 20 1 39 1 0 3 1 5 A Trench F Trench N % NISP % NISP 103/101 % NISP % Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Seals Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total 12.1 13.6 71.2 3.0 99.9 22.6 43.0 33.5 0.9 100.0 17.3 46.9 34.2 1.6 100.0 18 14.9 17 14.0 84 69.4 2 1.7 121 100.0 6 6.9 33 37.9 48 55.2 0 0.0 87 100.0 23 13.7 85 50.6 59 35.1 1 0.6 168 100.0 2 18 11.2 70 43.5 72 44.7 1 0.6 161 100.0 8 10.5 39 51.3 28 36.8 1 1.3 76 99.9 1 9 0 0 10 6 6.8 7 21 23.9 10 59 67.0 18 2 2.3 0 88 100.0 35 9 11.3 5 22 27.5 6 48 60.0 8 1 1.3 2 80 100.1 21 13 16.9 8 32 41.6 17 31 40.3 10 1 1.3 1 77 100.1 36 4 8 16 6 23 9 0 1 43 24 10 9.3 9 34 35.4 10 49 51.0 4 3 3.1 1 96 99.9 24 13 10.9 1 52 43.7 2 5445.40 00.00 119 100.0 11 15.5 31 43.7 29 40.8 0 0.0 71 100.0 3 4 4 17 17 12 12 1 1 34 34 Upper part: 0–10, 10–20, 20–30 Cattle Sheep and goat Seals Reindeer Total Lower part: 30–40, 40–50, etc. Cattle Sheep and goat Seals Reindeer Total 60 115 139 4 318 46 129 106 6 287 18.9 36.2 43.7 1.3 100.1 16.0 44.9 36.9 2.1 99.9 47 12.4 135 35.7 193 51.1 3 0.8 378 100.0 27 10.9 118 47.8 100 40.5 2 0.8 247 100.0 28 11.4 75 30.6 138 56.3 4 1.6 245 99.9 42 11.6 150 41.3 167 46.0 4 1.1 363 100.0 20 21.7 155 15.0 33 35.9 358 34.7 36 39.1 506 49.0 3 3.3 14 1.4 92 100.0 1033 100.1 Supplements. NISP counts. Square 100/102, level 276: Cattle, 2. Sheep and goat, 3. Seals, 2. Trench F: Sheep and goat, 2 18 35.3 18 35.3 13 25.5 2 3.9 51 100.0 133 14.0 415 43.8 386 40.7 14 1.5 948 100.0 51 2018 Journal of the North Atlantic G. Nyegaard Appendix 1 B. Distribution of animal bones in excavated trenches and layers. No. 37 Trench Upper part: 0–10, 10–20, 20–30, 30–40 NISP Cattle 7 Sheep and goat 14 Seals 54 Reindeer 5 Total 80 Lower part: 40–50, 50–60, etc. Cattle 34 Sheep and goat 84 Seals 89 Reindeer 2 Total 209 L % 8.8 17.5 67.5 6.3 100.1 16.3 40.2 42.6 1.0 100.1 52 2018 Appendix 1 C-1. Distribution of Journal of the North Atlantic G. Nyegaard animal bones in excavated trenches and layers. No. 37 The bog Layer 0–10 Layer 10–20 Layer 20–30 Layer 30–40 Layer 40–50 Layer 50–60 Layer 60–70 Upper part Lower part . Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Trench B NISP % 36 34.6 19 18.3 44 42.3 5 4.8 104 100.0 54 17.6 78 25.5 112 36.6 62 20.3 306 100.0 43 11.1 145 37.5 167 43.2 32 8.3 387 100.1 45 12.0 147 39.2 180 48.0 3 0.8 375 100.1 20 11.4 84 48.0 69 39.4 2 1.1 175 99.9 4 3 34 0 41 0 1 6 0 7 Trench C NISP % 24 19.8 44 36.4 51 42.1 2 1.7 121 100.0 39 15.4 133 52.6 76 30.0 5 2.0 253 100.0 74 24.2 113 36.9 118 38.6 1 0.3 306 100.0 139 19.4 314 43.8 249 34.7 15 2.1 717 100.0 22 16.3 63 46.7 50 37.0 0 0.0 135 100.0 Trench D NISP 0 5 2 1 8 14 12 23 0 49 31 60 65 5 161 62 131 163 98 454 58 132 135 68 393 19 50 43 0 112 1 1 1 0 3 Layer 0–10, NISP % 28.6 24.5 46.9 0.0 100.1 19.3 37.3 40.4 3.1 100.1 13.7 28.9 35.9 21.6 100.1 14.8 33.6 34.4 17.3 100.1 17.0 44.6 38.4 0.0 100.0 Layer 0–10, 10–20, 20–30 NISP % Cattle 133 16.7 Sheep and goat 242 30.4 Seals 323 40.5 Reindeer 99 12.4 Total 797 100.0 Layer 30–40, 40–50, etc. Cattle 69 11.5 Sheep and goat 235 39.3 Seals 289 48.3 Reindeer 5 0.8 Total 598 99.9 . %*) 19.1 34.7 46.3 100.1 *) excl. reindeer 10–20, 20–30, 30–40 % *) 18.8 36.6 44.5 99.9 17.7 41.6 40.7 100.0 *) excl. reindeer 137 20.1 290 42.6 245 36.0 8 1.2 680 99.9 161 18.9 377 44.2 299 35.1 15 1.8 852 100.0 % 107 15.9 208 31.0 253 37.6 104 15.5 672 100.0 Layer 40–50, etc. 78 15.4 183 36.0 179 35.2 68 13.4 508 100.1 53 2018 Journal of the North Atlantic G. Nyegaard Appendix 1 C-2. Distribution of animal bones in excavated trenches and layers. No. 37 The bog Layer 0-10 Layer 10-20 Layer 20-30 Layer 30-40 Layer 30-50 Layer 40-50 Layer 50-60 Layer 60-70 Upper part Lower part Trench Eb NISP Cattle 3 Sheep and goat 12 Seals 20 Reindeer 0 Total 35 Cattle 9 Sheep and goat 37 Seals 30 Reindeer 2 Total 78 Cattle 12 Sheep and goat 38 Seals 46 Reindeer 6 Total 102 Cattle 9 Sheep and goat 21 Seals 32 Reindeer 0 Total 62 Cattle Seals Reindeer Total Cattle 2 Sheep and goat 6 Seals 14 Reindeer 0 Total 22 Cattle 2 Sheep and goat 3 Seals 12 Reindeer 0 Total 17 Cattle 1 Sheep and goat 2 Seals 0 Reindeer 0 Total 3 Layer 0–10, 10–20, 20–30 Cattle 24 Sheep and goat 87 Seals 96 Reindeer 8 Total 215 Layer 30–40, 40–50, etc. Cattle 14 Sheep and goat 32 Seals 58 Reindeer 0 Total 104 % 11.5 47.4 38.5 2.6 100.0 11.8 37.3 45.1 5.9 100.1 14.5 33.9 51.6 0.0 100.0 Trench G Trench H NISP % NISP % 9 3 14 10 6 9 0 1 29 23 68 54.4 15 19.0 19 15.2 21 26.6 37 29.6 37 46.8 1 0.8 6 7.6 125 100.0 79 100.0 46 18.6 11 4.7 133 53.8 35 15.0 51 20.6 86 36.9 17 6.9 101 43.3 247 99.9 233 99.9 13 8.2 31 8.1 76 48.1 120 31.4 60 38.0 115 30.1 9 5.7 116 30.4 158 100.0 382 100.0 1 3 2 6 14 20.3 22 7.2 16 23.2 155 50.5 38 55.1 123 40.1 1 1.4 7 2.3 69 100.0 307 100.1 14 17.7 26 32.9 38 48.1 1 1.3 79 100.0 123 30.7 29 8.7 166 41.4 66 19.7 94 23.4 132 39.4 18 4.5 108 32.2 401 100.0 335 100.0 27 11.9 68 8.8 92 40.5 301 38.9 98 43.2 279 36.0 10 4.4 126 16.3 227 100.0 774 100.0 Trench I NISP % 1 0 1 0 2 7 6 2 0 15 4 10 17 0 31 15 16.1 29 31.2 29 31.2 20 21.5 93 100.0 33 17.9 84 45.7 52 28.3 15 8.2 184 100.1 7 21 27 1 56 Layer 0–10, 10–20, 20–30, 30–40 11.2 40.5 44.7 3.7 100.1 13.5 30.8 55.8 0.0 100.1 % *) 12.8 29.1 58.1 100.0 10.5 46.5 43.1 100.1 *) Tr. H. excl. reindeer 27 19.1 45 31.9 49 34.8 20 14.1 141 99.9 % *) 22.3 37.2 40.5 Layer 40–50, 50–60 40 16.7 105 43.8 79 32.9 16 6.7 240 100.1 17.9 46.9 35.3 100.1 *) Tr. I excl. reindeer 100.0 54 2018 Appendix 1 C-3. The bog Layer 0–10 Layer 10–20 Layer 20–30 Layer 30–40 Layer 40–50 Layer 50–60 Layer 60–70 "The trench" Upper part Lower part Journal of the North Atlantic G. Nyegaard Distribution of animal bones in excavated trenches and layers. No. 37 Trench Qa NISP % 3 5 9 0 17 9 10.1 30 33.7 50 56.2 0 0.0 89 100.0 16 10.1 63 39.9 74 46.8 5 3.2 158 100.0 31 15.9 51 26.2 109 55.9 4 2.1 195 100.1 57 16.2 139 39.4 155 43.9 2 0.6 353 100.1 12 15.8 38 50.0 25 32.9 1 1.3 76 100.0 1 1 5 7 5 21 17 1 44 NISP % 59 12.9 149 32.5 242 52.7 9 2.0 459 100.1 70 16.1 178 40.8 185 42.4 3 0.7 436 100.0 Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Trench K Trench P NISP % NISP % 0 4 3 13 13 7 0 1 16 25 13 17.3 12 19.0 18 24.0 25 39.7 43 57.3 24 38.1 1 1.3 2 3.2 75 99.9 63 100.0 38 16.9 18 16.7 69 30.7 19 17.6 115 51.1 45 41.7 3 1.3 26 24.1 225 100.0 108 100.1 51 14.7 27 8.6 217 62.7 56 17.9 75 21.7 106 33.9 3 0.9 124 39.6 346 100.0 313 100.0 7 10.6 21 13.1 31 47.0 56 35.0 28 42.4 62 38.8 0 0.0 21 13.1 66 100.0 160 100.0 0 9 9.6 0 46 48.9 0 38 40.4 0 1 1.1 0 94 100.1 1 1 1 1 2 2 Layer 0–10, 10–20, 20–30 NISP % Cattle 51 16.1 Sheep and goat 90 28.5 Seals 171 54.1 Reindeer 4 1.3 Total 316 100.0 Layer 30–40, 40–50, etc. Cattle 58 14.1 Sheep and goat 248 60.2 Seals 103 25.0 Reindeer 3 0.7 Total 412 100.0 Layer 0–10, 10–20, 20–30, 30–40 NISP % %*) 61 12.0 113 22.2 182 35.8 153 30.1 509 100.1 Layer 40–50, 50–60, 30 11.8 102 40.2 100 39.4 22 8.7 254 100.1 etc. 17.5 30.6 51.9 100.0 12.9 44.0 43.1 100.0 *) Tr. P excl. reindeer 55 2018 Journal of the North Atlantic G. Nyegaard Appendix 1 C-4. Distribution of animal bones in excavated trenches and layers. No. 37 The bog Layer 0–10 Layer 10–20 Layer 20–30 Layer 30–40 Layer 40–50 Layer 40–60 Layer 50–60 Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Trench Qb NISP 4 5 17 0 26 15 35 24 4 78 12 30 31 2 75 49 113 88 0 250 46 104 62 1 213 Trench S % NISP % 8 20.0 12 30.0 19 47.5 1 2.5 40 100.0 19.2 19 9.6 44.9 41 20.7 30.8 96 48.5 5.1 42 21.2 100.0 198 100.0 16.0 35 11.1 40.0 140 44.6 41.3 133 42.4 2.7 6 1.9 100.0 314 100.0 19.6 40 21.2 45.2 57 30.2 35.2 85 45.0 0.0 7 3.7 100.0 189 100.0 21.6 19 15.3 48.8 70 56.5 29.1 32 25.8 0.5 3 2.4 100.0 124 100.0 Cattle 2 Sheep and goat 5 Seals 2 Reindeer 1 Total 10 Cattle Sheep and goat Seals Reindeer 0 0 Total 26 23 6 6 11 12 9 5 Upper part: 0–10, 10–20, 20–30 Cattle Sheep and goat Seals Reindeer Total Lower part: 30–40, 40–50, etc. Cattle Sheep and goat Seals Reindeer Total NISP 31 70 72 6 179 103 233 161 2 499 % NISP % 17.3 62 11.2 39.1 193 35.0 40.2 248 44.9 3.4 49 8.9 100.0 552 100.0 20.6 65 19.3 46.7 139 41.4 32.3 122 36.3 0.4 10 3.0 100.0 336 100.0 Supplement. NISP counts. Square meter 105/96: Cattle, Seals, 34. Reindeer, 7. 23. Sheep and goat, 30. 56 2018 Journal of the North Atlantic G. Nyegaard Appendix 1 C-5. Distribution of animal bones in excavated trenches and layers. The mire. Summary of trenches B, C, D, Eb, G, H, I, K, P, Qa, Qb, S. No. 37 Trench O*) NISP % Cattle 44 21.6 Sheep and goat 95 46.6 Seals 56 27.5 Reindeer 9 4.4 Total 204 100.1 *) Square meter 109/102 was not excavated in 10 cm layers. The bones were documented with x, y, and z - coordinates. Supplement: NISP counts of bones from square meter 105/96 + trench K,P, and Qa: Cattle, 29. Sheep and goat, 53. Seals, 52. Reindeer, 8. Layer 0–10 Layer 10–20 Layer 20–30 Layer 30–40 Layer 40–50 Layer 50–60 Layer 60–70 Layer 30–50 Layer 40–bottom Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total NISP % 95 21.3 142 31.8 198 44.4 11 2.5 446 100.0 Excl. reindeer % 21.8 32.6 45.5 99.9 274 19.5 21.4 455 32.3 554 39.3 125 8.9 35.5 43.2 1408 100.0 100.1 Cattle 2 Sheep and goat 4 Seals 7 Reindeer 0 Total 13 Cattle 1 Seals 3 Reindeer 2 Total 6 Cattle 3 Sheep and goat 6 Seals 7 Reindeer 1 Total 17 340 14.5 855 36.4 948 40.4 204 8.7 15.9 39.9 44.2 2347 100.0 100.0 512 14.5 16.3 1332 37.7 42.5 1291 36.5 41.2 399 11.3 3534 100.1 100.0 321 14.6 15.4 940 42.7 45.2 820 37.3 39.4 120 5.5 2201 100.1 100.0 79 15.1 15.2 210 40.1 40.4 231 44.1 44.4 4 0.8 524 100.1 100.0 57 2018 Journal of the North Atlantic No. 37 Appendix 1 D. Distribution of The Norse ditch G. Nyegaard animal bones in excavated trenches and layers. Layer 0–10 Layer 10–20 Layer 20–30 Layer 30–40 Layer 40–50 Layer 50–60 Layer 60–70 Layer 70–80 L. 80–bottom Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Cattle Sheep and goat Seals Reindeer Total Trench Ea NISP % 9 13.4 14 20.9 43 64.2 1 1.5 67 100.0 7 12.7 21 38.2 23 41.8 4 7.3 55 100.0 30 20.7 51 35.3 60 41.4 4 2.8 145 100.1 44 21.6 72 35.3 85 41.7 3 1.5 204 100.1 33 17.3 83 43.5 71 37.2 4 2.1 191 100.1 36 22.6 76 47.8 44 27.7 3 1.9 159 100.0 15 16.3 39 42.4 38 41.3 0 0.0 92 100.0 22 21.6 29 28.4 46 45.1 5 4.9 102 100.0 40 26.7 48 32.0 58 38.7 4 2.7 150 100.1 Layer 0–10 Layer 10–20 Layer 20–30 Layer 30–40 Layer 40–50 Layer 50–bottom "The Trench" Trench M NISP % Cattle 5 Sheep and goat 7 Seals 32 Reindeer 0 Total 44 Cattle 8 10.5 Sheep and goat 20 26.3 Seals 44 57.9 Reindeer 4 5.2 Total 76 99.9 Cattle 14 12.8 Sheep and goat 47 43.1 Seals 46 42.2 Reindeer 2 1.8 Total 109 99.9 Cattle 28 31.1 Sheep and goat 29 32.2 Seals 31 34.4 Reindeer 2 2.2 Total 90 99.9 Cattle 11 Sheep and goat 10 Seals 16 Reindeer 0 Total 37 Cattle 33 36.3 Sheep and goat 19 20.9 Seals 33 36.3 Reindeer 6 6.6 Total 91 100.1 Cattle 25 Sheep and goat 18 Seals 41 Reindeer 3 Total 87 Supplement: Square 105/110. NISP counts: Cattle, 7. Sheep and goat, 26. Seals, 35. Reindeer, 1. Totals for Trench Ea NISP % Cattle Sheep and goat Seals 468 40.2 Reindeer 28 2.4 1165 100.1 Totals for Trench M NISP % Cattle 124 23.2 Sheep and goat 150 28.1 Seals 243 45.5 Reindeer 17 3.2 534 100.0 236 20.3 433 37.2 58 2018 Journal of the North Atlantic No. 37 G. Nyegaard Appendix 2. Survey of vertical distribution for bones of cattle, sheep and goat, seals, and reindeer trench by trench Several factors mean that the excavation layers cannot be directly compared chronologically and stratigraphi­ cally from trench to trench. The survey levels reveal that the thickness of the cut-over and considerably uneven peat surface varies by some centimetres from trench to trench. This results in minor differences in level with respect to the starting point for layer 0–10 and, consequently, all the subsequent 10 cm layers. However, the primary factor is that the terrain is not flat and the thickness of the peat deposits varies. In total, the terrain falls 45 cm along the x­axis of the coordinate system from trench N to trench G. In trench B there is accordingly a difference in level of up to 20 cm for several mire layers over the course of the short distance from (x, y) = (106, 102) to (x, y) = (108.5, 102) (Fig. 8D). Along the y-axis, there is a smaller fall in terrain of 10–15 cm from trench D towards trench E – a consequence of more rapid peat accumulation in the wetter trenches to the south. On dry land, the greatest concentration of animal bones is seen in trenches A and F from excavation layers 20–30 and 30–40. These correspond to layers 4 and 5 on the section drawing shown in Figure 8B-C. Closer to the dwelling, in trench N, the cultural deposits are deeper and the bones have a greater vertical distribution, extending down to layer 70–80 (Fig. 8A). The same trend in the change in species distribution is seen down through the layers in all three trenches (Ap­ pendix 1 A). The proportion of bones of sheep and goat is on average c. 9 % lower in the upper three layers compared with the underlying layers. Conversely, the proportion of seal bones is on average 8.3 % higher in the upper part relative to the layers below. With respect to cattle, there is a slight increase in the upper three layers in trenches A and F, whereas the proportion in both the upper and lower part of trench N is almost identical. No differences can be seen in the content of reindeer, which is insignificant in both the upper and the lower part of the deposits – on average 1.4 %. Appendix 1 also provide additional information with respect to differences in the horizontal distribution. For example, the summaries for square 103/101 and trenches A, F, and N show an increase in the proportion of seal bones when approaching the dwelling from the edge of the mire in both the upper and the lower deposits. For cattle and sheep/goat there is a tendency for this trend to proceed in the opposite direction. In the modest assemblage of bones from trench L (Appendix 1 B), a similar increase is seen in the content of seal in the upper four layers, whereas the proportion of sheep and goat declines. Cattle are less well represented in the upper layers than the lower. In the mire (Appendix 1 C), most of the animal bones from trench B were found equally distributed between layers 10–20, 20–30, 30–40 and 40–50, corresponding to the mire deposits located between levels 242 and 282 at coordinates (x, y) = (107, 102) and between levels 240 and 280 at coordinates (x, y) = (107, 100) on Figure 8D. The levels here refer to the distance below the theodolite’s line of sight in a fixed position. In trench C, there were the largest number of bones from the same four excava­ tion layers which, at coordinates (x, y) = (107, 104), lie between levels 246 and 284. In trench D, the two upper excavation layers, in contrast to trench B, contain only a very few bones. Conversely, there is a high concentration in layers 20–30, 30–40 and 40–50, which at (x, y) = (107, 98) lie between levels 243 and 272. Most come from layer 5 which corresponds to the find-rich layer 8 on Figure 8D. One half of trench E (trench Ea) is, as already mentioned, cut through by a ditch which continues into trench M. The fill of this ditch dates from a late phase of the occupation. Just less than three quarters of the bones from trench E come from squares (105/107), (105/108) and (106/108), originating almost exclusively from the ditch fill. The same is true of the bones from trench M. From the summary of the animal bones recovered from the mire, excluding trenches Ea and M, it is possible to identify some general trends (Appendix 1 C-5). Compari­ son of the content of layer 0–10 or layer 10–20 with the underlying layers suggests, as was the case with trenches A, F, and N, an overall relative decline in sheep and goat through the five uppermost excavation layers. The propor­ tion of seal bones is not as high in layer 0–10 as in the uppermost layers in the trenches on dry land, neither is the tendency for increasing content through the upper five layers as marked. In layer 50–60, seal bones make up just as great a proportion (44.1 %) as in the uppermost horizon (44.4 %). Cattle make up a greater proportion in the two upper layers (21.3 and 19.5 %) than in the deeper lying layers (14.5–15 %), i.e. the same trend which saw slightly less pronounced expression in trenches A and F. Reindeer bones occur in relatively large quantities in some of the find-rich layers in the middle of the mire section, whereas their proportion in the uppermost and lowermost finds ho­ rizons is of the same order as in for example the trenches on dry land. Evaluated trench by trench, the picture of the species distribution in the mire layers is more nuanced: Trench B corresponds approximately to the overall summary for the mire, apart from the relative decline in seal in the uppermost three layers compared with those below. However, this becomes virtually non-existent when reindeer bones are excluded from the calculation. The latter have particularly high numbers in layers 10–20 and 20–30. In trench C some of the same trends can be traced which were apparent in the other trenches, but an overall summary for the four animal groups in the three upper find layers does not differ significantly from the overall content of the two lower layers. In trench D there is a large content of reindeer bones in layers 30–40 and 40–50. These obscure the relative differences between the other species. If reindeer bones are excluded from the summary and the boundary for a vertical division is fixed between layers 30–40 and 40–50, such that more or less equal numbers of bones occur in both, the same trends are apparent from bottom to top as demonstrated in trenches A, F, and N, although slightly less distinct. The percentage composition in the upper parts of, respectively, trench D and trench A is accordingly 59 2018 Journal of the North Atlantic No. 37 G. Nyegaard virtually identical when reindeer bones are excluded from the calculation. In trench G, the total number of seal bones (43 frag­ ments) exceeds that for sheep and goat (33 fragments) in the two uppermost layers, whereas there are more sheep and goat bones (225 fragments) than seal bones (149 frag­ ments) in the three underlying layers. However, as seen in some of the other trenches in the mire, there is a greater number of bones of seal than of sheep and goat in the lowermost excavation layer. The modest assemblage has a high proportion of cattle remains in some of the layers, notably layer 10–20. With a vertical division of the mate­ rial between layer 20–30 and layer 30–40 there is instead an apparent decline in seals in the three upper layers. In trench H there is a relative decline in sheep and goat in the three upper layers relative to the layers below them, accompanied by an increase in seal. This tendency is particularly clear if the reindeer bones from layers 20–30 and 30–40 are excluded from the calculations. Once again, there are more bones of seal than of sheep and goat in the lowermost excavation layer. Cattle remains show rela­ tively low proportions in this trench. The summary for trench I reveals virtually the same picture as seen in trench H, apart from a greater content of cattle remains. Once again there are more bones of seal than of sheep and goat in the lowermost excavation layer. The summary for trench K similarly shows a decline in the relative content of sheep and goat in the three uppermost layers and a corresponding increase in seal. With respect to cattle there is a minor increase in the uppermost layers. In trench P, the content of the four uppermost layers is compared with that of the two lowermost. The same pat­ tern emerges yet again: A relative decline in sheep and goat and an increase in seal in the upper part, which most clearly sees expression when reindeer bones are excluded from the calculations. The bones from trench Q are, due the shape of the trench, summarised in two parts. In the four square me­ tres running alongside the drainage ditch (trench Qa), the content of the four upper layers, compared with that of the underlying three layers, suggests a relative reduction in cattle, a decline in sheep and goat and an increase in seal. The content of the remaining part of the trench (trench Qb) shows an overall relative reduction in cattle in the upper three layers, a decline in sheep and goat and an increase in seal. In trench S there is, as was the case in trench Q, a relative reduction in cattle in the three upper layers. Fur­ thermore, a decline is again seen in sheep and goat and an increase in seal relative to the overall summary for the lower layers. The small portion of bones recovered from the three square metres of trench E located outside the Norse ditch (trench Eb) is similarly summarised (Appendix 1 C-2). Bones of seal are slightly more numerous than in the other ditches. From coordinates (x, y) = (107, 109), in the sec­ tion recorded between trenches E and M (Fig. 10), there is a radiocarbon date for a peat sample taken between levels 300 and 320, corresponding to the level of the lowermost cultural deposits in this part of trench E. The date is AD 1030–1220 (cal. ± 2 standard deviations) (K-7051) and appears to represent the main settlement. The bones derive from layers which are later than this date. The animal bones from the three square metres of trench E which are cut through by the ditch (trench Ea) come from the final part of the occupation (Appendix 1 D), and the ditch fill must be presumed to have been deposited over a relatively short period of time. Most of trench M is cut through by the same ditch. It was investigated rather more extensively towards the end of the excavation in 1998. This could be the reason that sheep and goat make up a slightly smaller proportion than in trench Ea. The composition of the contents corresponds otherwise more or less to that of trench Ea. 60 2018 Journal of the North Atlantic G. Nyegaard Appendix 3. Summary of bones of sheep and goat. The numbers of identified fragments of each No. 37 Ea +M% 20 3.3 0 0.0 38 6.2 59 9.7 38 6.2 44 7.2 8 1.3 11 1.8 6 1.0 23 3.8 20 3.3 28 4.6 0 0.0 0 0.0 0 0.0 48 7.9 0 0.0 17 2.8 26 4.3 30 4.9 3 0.5 22 3.6 1 0.2 34 5.6 18 8 73 99 1.8 7 0.9 75 1.8 13 2.1 Femur 401011051.970.9812.0122.0 Patella Tibia Fibula Astragalus Calcaneum Tarsales Metatarsus Phalanx 1 Phanlax 2 Phalanx3 0 0 Ossa sesam. 0 0 Metapodium 3 11 13 13 0.2 137 184 3.3 5 5 0.1 25 100 1.8 29 83 1.5 51 51 0.9 243 316 5.6 194 194 3.4 68 68 1.2 32 32 0.6 0 0 0.0 137 151 2.7 4829 5633 100.1 8 0.2 132 3.2 0 0.0 74 1.8 62 1.5 34 0.8 245 6.0 147 3.6 48 1.2 21 0.5 0 0.0 95 2.3 4092 100.1 2 0.3 19 3.1 1 0.2 16 2.6 4 0.7 3 0.5 23 3.8 15 2.5 3 0.5 5 0.8 0 0.0 17 2.8 609 100.1 Ovis Capra Ovis/Capra Ovis+Capra % 13 78 1.4 A+F+N % 4 0.5 2 0.3 56 7.2 68 8.7 55 7.1 89 11.4 13 1.7 6 0.8 6 0.8 20 2.6 14 1.8 19 2.4 2 0.3 0 0.0 0 0.0 3 0.4 28 3.6 4 0.5 9 1.2 16 2.1 12 1.5 38 4.9 31 4.0 17 2.2 5 0.6 0 0.0 32 4.1 778 100.3 skeletal element are shown. The mire % 51 1.2 14 0.3 302 7.4 433 10.6 281 6.9 409 10.0 27 0.7 26 0.6 39 1.0 Proc. cornu. Hyoid Calvarium Dentes sup. Mandibula Dentes inf. Dentes Atlas 41 24 0 0 40 28 0 0 27 26 9 10 0 0 5 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 17 10 48 31 33 13 1 18 22 13 0 0 51 40 17 17 0.3 346 414 7.3 573 573 10.2 331 384 6.8 536 555 9.9 Axis Ver. cer. 3–7 Ver. thorac. Ver. lumbal. Ver. sacral. Ver.caudal. Ver. indet. Costa Sternum Scapula Humerus Radius Radius-ulna Ulna Carpus Metacarpus Pelvis 49 49 0.9 38 47 0.8 52 52 0.9 0 0 32 15 0 0 37 38 27 27 0 0 39 34 0 0 0 0 163 163 2.9 163 163 2.9 133 133 2.4 115 2.8 126 3.1 80 2.0 9 0.2 0 0.0 1 0.0 383 9.4 0 0.0 76 1.9 128 3.1 184 4.5 35 0.9 73 1.8 29 0.7 249 6.1 11 11 0.2 0 0 0.0 1 1 0.0 501 501 8.9 0 0 0.0 78 105 1.9 108 187 3.3 213 259 4.6 23 42 0.7 82 117 2.1 36 36 0.6 254 345 6.1 62 8.0 0 0.0 12 1.5 28 3.6 37 4.8 2 0.3 17 2.2 6 0.8 51 6.6 Total 454 350 A+F+N: Include Square 103/101. The mire: Trenches B, C, D, Eb, G, H, I, K, O, P, Qa, Qb, and R. Ea + M: Include a small number of bones from Square 105/110. 61 2018 Journal of the North Atlantic G. Nyegaard Appendix 4. Calculation of “Minimum Numbers of Individuals” (MNI). No. 37 Caribou, trench H + P dextra sinistra Maxilla (molar teeth) Mandibula (molar teeth) Scapula Humerus,proximal 1 0 6 4 1 0 Humerus, distal 10 6 56 57510 Radius, proximal 75 26 30 6 8 Radius,distal 71016127211 Ulna 8835392611 Radius-ulna, proximal 0014110000 Radius-ulna, distal 00240000 Carpus 87783422 dextra sinistra Cattle Sheep + Goat dextra sinistra Caribou dextra sinistra 11 13 9 10 Metacarpus, proximal 6 5 50 Metacarpus, distal 2 8 23 Pelvis 5 3 17 Femur, proximal 0 2 7 Femur, distal 0 0 8 Tibia, proximal 2 2 10 Tibia, distal 8 6 46 Astragalus 8 11 44 Calcaneum 9 12 36 Tarsus,excl. astrag. and calca. 9 10 24 Metatarsus, proximal 8 9 34 Metatarsus, distal 7 6 12 *): dextra + sinistra 10 8 79 7365 91 831311 23 30 5 6 2 2 0 0 1 1 0 4 43 2 2 0 0 17 4*) 2*) 23 361 0 20 00 520 0 0 0 0 0 0 4 6 3 2 3 4 2 1 2 11 10 39 49 32 18 32 22 712 910 11 12 84 32 2 7*) 4*) 62 2018 Journal of the North Atlantic G. Nyegaard No. 37 Total NISP % 17 0.7 15 0.6 161 6.6 167 6.8 117 4.8 241 9.8 90 3.7 6 0.2 8 0.3 21 0.9 37 1.5 24 1.0 7 0.3 2 0.1 300 12.2 0 0.0 50 2.0 50 2.0 85 3.5 36 1.5 61 2.5 89 3.6 64 2.6 63 2.6 9 0.4 91 3.7 5 0.2 23 0.9 30 1.2 39 1.6 94 3.8 161 6.6 115 4.7 112 4.6 19 0.8 44 1.8 2453 100.1 Appendix 5. Summary of cattle bones. The numbers of identified fragments of each skeletal element are shown. A + F + N NISP % The mire NISP % 7 0.4 15 0.9 96 5.6 121 7.1 73 4.3 192 11.3 77 4.5 4 0.2 5 0.3 12 0.7 26 1.5 19 1.1 7 0.4 2 0.1 222 13.1 0 0.0 30 1.8 34 2.0 55 3.2 24 1.4 41 2.4 57 3.4 39 2.3 42 2.5 8 0.5 63 3.7 4 0.2 20 1.2 20 1.2 21 1.2 55 3.2 104 6.1 77 4.5 83 4.9 13 0.8 32 1.9 1700 99.9 The mire (a) NISP % The mire (b) NISP % Ea + M NISP % 8 2.2 0 0.0 26 7.1 27 7.4 20 5.4 17 4.6 2 0.5 2 0.5 2 0.5 6 1.6 7 1.9 4 1.1 0 0.0 0 0.0 49 13.3 0 0.0 13 3.5 10 2.7 18 4.9 7 1.9 6 1.6 15 4.1 14 3.8 11 3.0 1 0.3 15 4.1 0 0.0 2 0.5 5 1.4 8 2.2 19 5.2 23 6.3 16 4.4 8 2.2 3 0.8 3 0.8 367 99.8 Proc. cornualis Hyoid Clavarium Dentes superiores Mandibula Dentes inferiores Dentes 0 0.0 0.0 11.0 6.2 5.2 9.0 2.4 0.0 0.3 0.3 1.0 0.3 0.0 0.0 5.2 0.0 1.4 1.4 2.4 0.7 4.1 3.8 3.1 2.8 0.0 3.1 0.3 0.3 1.4 3.4 5.9 9.3 6.6 5.5 1.0 2.4 99.8 3 0 12 44 8 62 20 0 0 0 1 0 0 0 43 0 7 8 13 1 7 13 6 13 1 15 1 4 3 4 19 22 21 17 0 1 369 0.8 4 0.0 15 3.3 81 11.9 74 2.2 65 16.8 115 5.4 56 0.0 3 0.0 5 0.0 12 0.3 25 0.0 18 0.0 7 0.0 2 11.7 173 0.0 0 1.9 21 2.2 26 3.5 34 0.3 19 1.9 30 3.5 44 1.6 31 3.5 28 0.3 7 4.1 47 0.3 3 1.1 13 0.8 17 1.1 15 5.1 33 6.0 75 5.7 51 4.6 62 0.0 12 0.3 31 100.0 1254 0.3 1.2 6.5 5.9 5.2 9.2 4.5 0.2 0.4 1.0 2.0 1.4 0.6 0.2 13.8 0.0 1.7 2.1 2.7 1.5 2.4 3.5 2.5 2.2 0.6 3.7 0.2 1.0 1.4 1.2 2.6 6.0 4.1 4.9 1.0 2.5 100.2 0 32 18 15 26 7 0 1 1 3 1 0 Ver.caudales 0 Atlas Axis Ver. cerv. 3–7 Ver. thoracales Ver. lumbales Ver.sacrales Costa Sternum Scapula Humerus Radius Ulna Carpus Metacarpus 11 Pelvis Femur Patella Tibia Fibula Astragalus Calcaneum 4 Other tarsales Metatarsus Phalanx 1 Phalanx 2 Phalanx 3 Ossa sesamoides Metapodium 7 Total 290 15 0 4 4 7 2 12 10 17 27 19 16 9 8 0 9 1 1 3 A+F+N: Trench A, F, and The mire: Trenches B, C, D, Eb, G, H, I, K, O, P, Qa, Qb, and R. The mire (a): Sum of layer 0–10 and 10–20. The mire (b): Sum of layer 20–30, 30–40, etc. Trench Ea + M: Include a small number of bones from Square 105/110. N + Square 103/101. 63 2018 Journal of the North Atlantic G. Nyegaard Appendix 6. Specification of bones of pig, horse, dog, and cat. No. 37 Hyoid Calvarium Dentes sup. Mandibula Dentes inf. Dentes Atlas Axis Ver. cer. 3–7 Ver. thorac. Ver. lumbal. Ver. sacral. Ver. caudal. Costa Sternum Scapula Humerus Radius Ulna Carpus Metacarpus Pelvis Femur Patella Tibia Fibula Astragalus Calcaneum Tarsales Metatarsus Phalanx1 Phalanx2 Phalanx3 Phalanges Ossa sesam. Metapodium Felis catus 0 3 0 2 0 0 Sus Equus Canis domesticus caballus familiaris 0 0 0 9 1 2 0 3 0 7 0 2 6 5 2 0 6 3 1 0 0 2 0 0 1 0 1 9 2 3 4 0 4 0 0 0 0 0 0 0 0 2 1 0 0 2 0 1 2 1 1 2 1 2 1 0 0 2 1 0 1 0 0 2 1 1 1 0 1 0 0 0 3 0 2 1 0 0 1 2 0 0 2 2 1 0 0 4 1 0 6 0 1 2 1 0 2 1 0 1 0 0 0 0 0 2 0 0 Canis sp. 1 0 1 2 0 1 0 0 1 0 0 0 1 0 0 0 1 4 0 3 0 0 1 0 1 0 0 0 0 0 0 Total 76 28 30 1 21 64 2018 Journal of the North Atlantic G. Nyegaard Appendix 7. Specification of seal bones. The numbers of identified bone fragments of each skeletal element are shown. No. 37 % 0.0 8.2 2.5 4.3 0.8 0.5 1.8 3.3 1.2 0.5 0.3 0.6 26.9 0.6 2.6 2.9 2.8 2.6 1.0 2.7 2.3 2.0 0.3 2.7 2.1 0.2 0.7 0.6 1.3 6.0 3.9 9.8 1.7 0.3 100.0 Total 5562 6 8 197 15 Seal P. vitulina P. hispida 1 1 3 3 P. groenlandica 63 52 2 41 21 18 E. barbatus 4 3 6 1 1 C. cristata 57 37 25 5 79 2 6 8 2 10 1 2 10 1 3 3 2 3 1 6 2 3 268 All seals 1 498 153 258 48 31 111 197 71 29 16 39 1630 38 157 176 172 155 63 164 141 119 16 165 127 13 41 38 79 362 236 594 101 17 6056 Hyoid 1 Calvarium 369 Mandibula 59 1 Dentes 233 Atlas 48 Axis 31 Ver. cer. 3–7 111 Ver. thorac. 197 Ver. lumbal. 71 Ver. sacral. 29 Ver. caudal. 16 Ver. indet. 34 Costa 1551 Sternum 36 Scapula 149 Humerus 118 Radius 170 Ulna 120 Carpus 63 Metacarpus 163 Pelvis 139 Femur 89 Patella 16 Tibia 164 Fibula 124 Tibia-fibula 10 Astragalus 39 Calcaneum 35 Tarsales 78 Metatarsus 356 Phalanges, manus 234 Phalanges, pedis 591 Phalanges 101 Metapodium 17 4 1 65 2018 Journal of the North Atlantic No. 37 G. Nyegaard Appendix 8. Specification of bones of reindeer, walrus, polar bear, arctic fox and arctic hare. The numbers of identified bone fragments of each skeletal element are shown. Cornus Hyoid Calvarium Dentes sup. Mandibula Dentes inf. Dentes Atlas 3 0.3 29 5 0.5 6 0.6 2 0.2 Femur Patella Tibia Fibula Astragalus Calcaneum Tarsales Metatarsus Phalanx 1 Phalanx 2 Phalanx 3 Ossa sesam. Metapodium Total 51 5.2 31 3.2 6 0.6 4 66 6.8 6 0.6 19 1.9 34 3.5 1 19 1.9 73 7.5 23 2.4 18 1.8 5 0.5 0 0.0 29 3.0 975 99.8 48 11 Rangifer tarandus % 37 3.8 0 0.0 19 1.9 9 56 5.7 32 3.3 1 99 10.2 Axis Ver. cer. 3–7 Ver. thorac. Ver. lumbal. Ver. sacral. Ver. caudal. Costa Sternum Scapula Humerus Radius Ulna Radius-ulna Carpus Metacarpus Pelvis Os penis 1 2 0.2 9 0.9 1 0.1 0 0.0 119 12.2 2 1 0.1 22 2.3 49 5.0 56 5.7 14 1.4 19 1.9 22 2.3 1 22 2.3 1 1 1 1 2 19 Odobenus rosmarus Ursus maritimus 3 7 3 Alopex arcticus 2 2 1 6 1 1 1 1 3 2 2 1 23 Lepus arcticus 2 1 1 4 1 1 1 66 2018 Journal of the North Atlantic No. 37 Appendix 9. Specification of bird bones. Mallard White-tailed eagle 1 1 G. Nyegaard Ptarmigan 1 2 7 1 2 1 Guillemot/ Brünnich's guillemot Razorbill/Guillemot/ Brünnich's guillemot Calvarium Mandibula Vertebrae Costa Furcula Coracoid Scapula Sternum Humerus Radius Ulna Carpometacarpus Carpus Phalanges (man.) Pelvis Femur Tibiotarsus Fibula Tarsometatarsus Phalanges (ped.) 1 23 37 3 46 0 30 1 10 62 41 2 2 Total 1 2 14 129 92 67 2018 Journal of the North Atlantic G. Nyegaard Appendix 10. Measurements of animal bones. NB: All bones have been measuered in a dry state. * Indicates a slightly damaged bone. No. 37 Bos taurus Scapula Find no. Dextra x 2339 x 677 x 1666 x 2335 x 1326 x 1529 x 144 x 1681 Sinistra x 146 x 2443 x 893 x 1601 x 1326 x 162 x 2353 x 554 From 2001 x 516 x 522 x 141 x 165 Radius Find no. Dextra x 807 x 2371 From 2001 x 1486 x 263 x 1039 From 2001 x 1331 x 1453 x 2430 Sinistra From 2001 x 1647 x 2430 x 146 x 503 x 1142 x 777 x 138 x 2335 x 1402 x 1618 Metacarpus Find no. Dextra x 162 x 162 x 149 x 1400 x 288 x 1252 Sinistra x 895 x 151 x 149 x 2460 x 1198 x 832 x 162 x 1415 Square Layer SLC GLP 109/98 30-40 44.0 57.5 108/106 10-20 44.1 59.5 110/99 20-30 40.7 50.6 103/99 10-20 44.1 58.2 105/109 30-40 50.4 106/109 50-60 44.4 66.0 Stray find 42.7* 56.3 105/108 30-40 44.2 Stray find 58.1 109/96 20-30 46.7 59.4* 108/96 30-40 47.0 64.7 105/108 50-60 46.1* 59.5 105/109 30-40 48.5 62.4* Stray find 53.8 72.1 109/97 30-40 64.3 Stray find 56.9 69.2 Stray find 43.3 59.4 106/100 30-40 108/104 30-40 54.8 Stray find 46.9 64.4 Stray find 55.7 Square Layer GL Bp 105/100 30-40 236.0* 66.8 109/104 30-40 65.2 Stray find 75.7 105/107 20-30 63.1 106/105 20-30 73.7 98/101 0-10 76.3 109/103 30-40 71.3 103/105 30-40 68.2 105/109 Bottom layer 70.7* 109/103 20-30 Stray find 73.7 110/101 20-30 70.0 109/103 20-30 65.3 Stray find 69.0* 108/105 40-50 75.7 107/99 30-40 72.2 106/97 40-50 Stray find 103/99 10-20 110/100 20-30 105/107 90-bottom Square Layer GL Bp Stray find 181.6 Stray find 173.6 55.6 Stray find 52.9 106/102 30-40 53.5 107/105 30-40 48.4* 106/99 Niv 282 181.9 54.0 108/96 30-40 173.4 49.3* Stray find 166.2 54.0 Stray find 167.6 109/104 40-50 175.7 50.4 103/105 40-50 179.3 50.0* 106/96 40-50 54.5 Stray find 51.5 105/104 40-50 58.6 BG 38.7 41.6 36.3* 41.8 36.6 43.0 41.2 41.5 41.1* 44.5* 44.3 50.6 45.6 44.5* 39.8 43.0 41.3 45.5 39.2 BFp SD Bd 61.3 34.1 60.0 59.2 68.8 58.6 69.5 69.3 67.0 61.2 65.2 59.8 61.7* 69.4 65.6 SD Bd 29.1 52.8 33.2 53.4 31.2 32.6 53.8 30.0 51.9 31.7 55.7 27.2 45.0* 30.5 52.8 25.1* 52.0 54.8 63.0 72.1 68.7 52.0 60.7 68 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. x 165 x 146 x 151 x 1253 Dex. / Sin. x 2421 x 1474 x 764 From 2001 x 144 x 162 x 2460 x 929 x 1453 x 1474 x 1461 Tibia Find no. Dextra x 162 x 151 x 2413 x 147 x 2332 x 1618 x 335 x 1487 x 144 x 1567 x 544 x 137 Sinistra x 1522 x 2377 x 151 x 146 x 146 x 649 x 894 x 151 x 144 Astragalus Find no. Dextra x 149 From 2001 From 2001 x 163 x 299 x 879 x 2413 x 2377 Sinista No no. x 2413 x 142 x 2415 x 1670 x 1530 x 650 x 2439 x 135 x 1193 x 2396 x 2362 x 1615 x 221 x 1548 Stray find 52.4 Stray find 52.5 Stray find 50.3 107/99 50-60 107/95 30-40 105/107 30-40 103/101 20-30 Stray find Stray find Stray find 109/104 40-50 107/96 40-50 105/109 bottom layer 105/107 30-40 103/105 50-60 164.9 51.3 29.6 51.2 Square Layer Bd Stray find Stray find 105/110 70-80 54.5 Stray find 107/106 30-40 45.8 105/107 90-100 50.1 107/101 30-40 50.6 105/107 0-10 52.2 54.3 54.2 56.0 54.7 57.2 52.6 46.2 55.6 47.0 54.0 48.2 54.6 51.6 Stray find 109/100 40-50 52.1 106/97 30-40 53.8 Stray find 61.3* 105/109 Bottom layer 59.6 100/100 10-20 56.4 Stray find Stray find Stray find 106/96 30-40 47.1 108/96 40-50 54.5 Stray find Stray find Square Layer Stray find Stray find Stray find Stray find 108/105 30-40 108/97 40-50 From 2001 From 2001 Stray find From 2001 Stray find From 2001 107/103 10-20 106/98 40-50 107/97 30-40 105/110 50-60 Stray find 93/101 40-50 From 2001 From 2001 110/99 40-50 107/104 20-30 107/98 30-40 56.5 57.3 GLl GLm 54.3 49.6* 54.4 48.2 57.6 53.7 57.3 51.6 55.6 51.3 55.0 49.5 57.0 52.6 54.1 50.4 59.7 54.9 56.0 50.0 57.9 53.3 57.9 52.4* 54.3 50.2 55.5 51.8 58.6 53.4 60.9 55.7 55.9 51.8 60.7 54.5 60.9 55.2 56.3 51.4* 56.3 52.7 53.0 48.3 58.8 54.2 Bd 34.2 32.2* 34.6 37.9 38.5 34.0 35.2 37.0 41.3 35.5 38.5 35.9* 33.5 35.2 35.5 40.1 37.2 35.2 38.1 37.3* 34.8 34.4 39.3 48.8 53.4 50.5 48.9 54.2 69 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. Metatarsus Find no. Dextra x 644 x 151 x 1483 x 2395 x 2346 Sinistra x 1410 x 162 x 147 x 603 x 1649 x 786 Dex. /Sin. x 1426 x 139 x 147 x 728 x 585 x 2342 x 1398 x 554 x 151 x 535 x 162 Ovis aries Scapula Find no. Dextra x 2440 x 1401 x 1423 x 1497 x 1547 x 1550 x 1634 x 1664 x 744 Sinistra x 1620 x 2376 No no. x 785 x 873 x 1050 x 1429 x 1530 Humerus Find no. Dextra x 512 x 803 x 1404 x 1450 x 1649 x 1459 x 1399 x 1626 x 544 x 1324 x 1659 x 1640 x 1616 x 1647 x 1002 x 544 x 139 x 148 x 148 Square Layer 107/97 30-40 Stray find 107/107 20-30 105/110 10-20 103/97 30-40 GL 196.4* 204.4* Bp 42.7* 41.1 43.5 44.0 43.3 SD Bd 25.2 46.8* 21.4 47.6 22.8 43.1 52.6 48.2 51.5 52.1 44.1 45.9 50.6 45.6 45.6 46.6 SLC GLP 17.9 30.6 19.2 30.2 17.4 29.0 21.7 34.5 21.5 37.5 106/109 30-40 Stray find Stray find 41.3 102/101 30-40 39.7 105/107 50-60 38.4 103/101 40-50 41.4 107/103 10-20 Stray find Stray find 106/105 30-40 105/101 10-20 106/95 20-30 Stray find Stray find Stray find 108/104 10-20 Stray find Square Layer HS DHA 109/104 40-50 135.5 141.5 106/99 40-50 105/109 10-20 106/108 40-50 107/98 30-40 107/98 40-50 106/108 40-50 107/99 40-50 108/100 30-40 106/102 20-30 103/102 10-20 108/101 30-40 103/101 30-40 105/101 40-50 99/101 20-30 93/101 60-70 106/98 40-50 Square Layer Bd BT 108/100 40-50 26.6 24.7 106/104 20-30 29.2 27.2 94/101 60-70 36.3*** 33.1 106/98 20-30 28.0 25.8 105/107 50-60 26.4 25.3 105/99 20-30 32.6* 29.3 106/102 30-40 28.9 28.0 95/101 20-30 33.3 106/97 30-40 28.9 26.2 108/102 40-50 28.0 25.2 106/108 10-20 28.3 25.6* 109/101 20-30 29.3 27.4 106/97 50-60 32.6 29.6 110/101 20-30 30.4 27.9 108/99 10-20 26.4 24.6 106/97 30-40 26.5 24.2 Stray find 34.5* 31.1 Stray find 31.6* Stray find 27.2 25.5 BG 19.8 19.4 17.7 22.0 22.9 188.4 48.2*** 45.6 *** exostosis formation 20.5 32.8 19.7 33.3 19.8 33.1* 20.8 33.6 19.6 32.7 18.3 32.2 22.0 34.5 20.7 32.4 18.7 31.3 23.7 19.2 31.3 *** exostosis formation 22.0 22.5 20.0 20.6 19.6 19.6 21.0 20.9 34.4 22.9 70 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. x 148 x 2450 x 134 x 1487 x 1328 Sinistra x 1429 x 1459 x 1532 x 1398 x 1492 x 897 x 1586 x 832 x 1663 x 494 x 522 x 1639 x 1028 x 873 x 763 x 1524 x 1462 x 1643 x 1657 x 1628 x 148 x 2430 x 152 x 148 x 148 x 2432 x 2426 Radius Find no. Dextra x 148 x 605 x 264 x 1414 x 1649 x 1461 x 1451 x 858 x 148 x 152 x 606 x 937 From 2001 x 2348 Sinistra x 1568 x 1396 x 1005 x 1528 x 1057 x 1622 x 405 x 1652 x 2456 x 879 x 897 x 1643 x 148 x 1411 Metacarpus Find no. Dextra x 1000 x 146 x 1652 x 646 x 1622 Stray find 30.6* 28.2 103/108 0-10 26.1 Stray find 26.8 25.4 105/107 0-10 25.4 24.1 105/109 30-40 27.6 25.6 93/101 60-70 27.4 26.5 105/99 20-30 26.7 24.9 105/96 0-70 34.6 31.5 Stray find 29.7 26.1 109/100 30-40 29.2 27.0 107/96 30-40 31.1 29.6 97/101 20-30 32.2 28.0 106/96 40-50 25.5 23.9 107/99 40-50 31.5* 27.5 100/100 30-40 30.7 28.9 108/104 30-40 29.0 27.2 105/107 90-100 32.8 31.2 99/101 30-40 30.7 29.0 105/101 40-50 31.2 27.9 106/97 40-50 30.7 27.9 Stray find 27.5 93/101 30-40 29.6* 109/102 26.0 24.7 106/108 0-10 27.7 25.2 106/108 20-30 28.9 26.9 Stray find 27.2 25.9 109/103 20-30 29.5 26.2 Stray find 32.5 30.6 Stray find 29.7 27.6 Stray find 25.3 24.0 From 2001 24.8 23.8 From 2001 27.3 25.8 Square Layer GL BP Stray find 149.9 28.3 106/100 20-30 29.7 106/104 30-40 29.1 106/99 40-50 28.1 105/107 50-60 28.1 103/105 50-60 29.8 106/98 30-40 31.0* 103/101 0-10 30.7 Stray find Stray find 106/100 20-30 107/97 50-60 Stray find 109/95 20-30 24.6 107/108 20-30 139.6 29.2 106/109 20-30 135.4 26.3* 110/99 30-40 28.4 105/98 20-30 31.6 Trench F ? 34.1 105/107 80-90 28.5 108/101 0-10 27.8 105/107 40-50 103/108 10-20 30.5 108/97 40-50 29.4 107/96 30-40 29.3 106/99 40-50 Stray find 94/101 20-30 32.4 Square Layer GL Bp 110/100 40-50 115.7 21.6 Stray find 21.6** 105/107 40-50 22.0** 101/101 30-40 21.7** 105/107 80-90 20.0** BFp 26.3 26.4 26.2 15.0 26.4 16.0 27.0 27.9 22.8 26.0 25.9 26.7 15.3 27.0 32.7 26.0 25.4 29.0 27.7 30.6 SD Bd 13.7 25.6 13.6** 11.1** SD BD BFd 27.3 21.4 28.4 23.2 28.4 22.2 25.6 20.4 29.6 24.2 28.8* 25.7 25.3* 30.5 26.5 28.0 ** juv. ** juv. ** juv. ** juv. 15.1 14.9 14.9 71 2018 Journal of the North Atlantic No. 37 Bp 39.4 45.1 39.6 38.9 Bp 37.2 40.2 Bd 25.1 25.8 25.4 27.5 24.2 23.4 25.0 25.4 23.4* 24.0 26.7 23.8 24.5 25.7 26.8 23.8 21.4 G. Nyegaard Appendix 10, Continued. x 1411 x 1400 x 571 x 144 x 522 x 2440 x 1449 x 1399 x 1399 x 1399 Sinistra x 1528 x 858 x 1560 x 1461 x 863 x 1661 x 1674 x 139 Dex. / Sin. x 1661 x 1520 x 1823 x 588 x 858 x 1406 x 1533 x 1198 x 1403 x 1399 x 1477 x 1473 x 1520 x 276 x 1473 x 1034 x 494 x 1476 x 2440 x 148 x 2436 x 139 x 148 Femur Find no. Dextra x 807 Sinistra x 1561 x 1403 x 1050 Tibia Find no. Dextra x 937 x 1634 x 490 x 1572 x 148 x 148 x 937 x 721 x 166 x 1634 x 2330 x 1645 x 1479 x 2455 x 886 x 1463 x 1508 x 1659 x 148 94/101 20-30 106/102 30-40 102/100 30-40 Stray find 108/104 30-40 109/104 40-50 105/103 40-50 106/102 30-40 106/102 30-40 106/102 30-40 105/98 20-30 103/101 0-10 106/103 30-40 103/105 50-60 105/101 40-50 106/103 40-50 105/103 20-30 Stray find 106/103 40-50 105/103 40-50 106/101 40-50 107/105 30-40 103/101 0-10 108/99 40-50 106/98 50-60 103/105 40-50 106/102 30-40 106/102 30-40 105/98 20-30 106/103 40-50 105/103 40-50 106/105 40-50 106/103 40-50 108/99 20-30 100/100 30-40 105/97 30-40 109/104 40-50 Stray find From 2001 Stray find Stray find Square Layer 105/100 30-40 Trench E 106/102 30-40 99/101 20-30 Square Layer 107/97 50-60 106/108 40-50 101/100 30-40 107/108 50-60 Stray find Stray find 107/97 50-60 106/104 10-20 Stray find 106/108 40-50 From 2001 105/98 10-20 105/105 30-40 109/105 50-60 108/96 0-10 105/97 20-30 108/103 30-40 106/108 10-20 Stray find 134.8* 126.5 109.1 23.0* 24.1 22.3 21.4 23.2 22.7 22.0 22.4 22.9 22.2* 21.4 18.4 21.4** 22.2 22.0 23.4 22.7 20.9 15.0 24.7 11.4 22.3* 24.2 24.9 26.2 26.7 26.5 25.1 23.8 29.2 26.9 23.8 22.5* 24.4 26.6 25.7 21.8 23.5 24.5 23.0* 27.5 25.3 25.9 27.2 28.9 ** juv. 72 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. x 763 x 2341 x 1672 Sinistra x 1642 x 1199 x 763 x 335 From 2001 x 1457 x 487 x 1620 x 2429 x 1580 x 832 x 665 From 2001 x 2436 x 1429 x 193 x 145 x 2335 x 484 x 863 x 148 x 148 x 534 Astragalus Find no. Dextra x 1403 x 1403 x 604 x 1571 x 1472 x 2434 No no. x 2453 x 1502 x 1564 x 148 x 1452 x 1399 x 148 x 148 x 1449 x 168 x 1633 x 140 Sinistra x 1492 x 897 x 753 x 138 x 142 x 2440 x 863 x 167 x 1578 x 2430 x 228 x 588 x 867 x 335 x 143 x 405 x 842 x 264 x 1586 x 1553 x 484 106/97 40-50 26.8 From 2001 23.1 97/101 10-20 23.2 106/99 40-50 24.9 103/105 60-70 26.0 106/97 40-50 23.9 107/101 30-40 26.3 Stray find 24.2 108/98 30-40 21.7 107/100 10-20 21.7 106/102 20-30 24.6* 105/110 50-60 26.1 107/107 10-20 25.9 106/96 40-50 24.3 106/96 30-40 23.2 109/103 30-40 24.8 From 2001 23.4 93/101 60-70 24.1 101/100 20-30 23.7 Stray find 26.5 103/99 10-20 24.3 106/100 10-20 22.6 105/101 40-50 24.6* Stray find 27.1 Stray find 24.3 106/100 40-50 26.5 Square Layer GLl GLm 106/102 30-40 26.6 25.9 106/102 30-40 24.3 24.1 106/105 10-20 25.8 24.4 105/108 40-50 24.5 24.0 106/109 10-20 25.9 25.5 From 2001 26.8 25.9 108/101 30-40 24.6 24.2 From 2001 25.2 24.6 106/109 40-50 23.8 22.6 106/108 30-40 24.4 23.6 Stray find 105/109 Bottom layer 25.5 25.5 106/102 30-40 25.0 24.8 Stray find 24.5 24.7 Stray find 25.7 25.1 105/103 40-50 26.8 25.0 Stray find 27.2 107/107 30-40 25.3 25.5 Stray find 25.1 25.4 109/100 30-40 24.3 24.0 107/96 30-40 26.4* 25.4* 105/100 20-30 29.9* 29.4 Stray find 27.8* 27.1 Stray find 25.9 26.1 From 2001 24.7* 24.6 105/101 40-50 25.2 24.0 Stray find 26.8 26.7 107/107 40-50 25.0 25.1 From 2001 25.8 25.2 107/105 30-40 25.3 24.8 107/105 30-40 24.7* 24.6 105/101 40-50 25.5 24.4 107/101 30-40 24.8 25.1 Stray find 29.4 29.2 108/101 0-10 25.7 25.7 108/100 60-70 23.7* 23.5 106/104 30-40 97/101 20-30 22.7 21.8 105/104 50-60 23.5 23.2 106/100 10-20 23.8 22.9 Bd 16.4 15.2 16.1 16.3 16.0 17.6 16.8 16.7 16.0 15.1 16.5 16.6 16.3 17.1 16.7 17.8 17.5 16.6 16.6 15.7 17.6 20.6 16.2 16.5 16.4 16.3 16.9 15.6 15.2 16.0 16.1 17.0 19.7 15.8 15.8 17.3 15.0 15.2 16.7 73 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. Calcaneum Find no. Dextra x 138 x 138 x 1560 x 1046 x 1193 x 1663 x 1628 x 518 x 335 x 1412 x 2340 x 786 Sinistra x 2430 x 289 x 753 x 152 x 720 x 376 x 1411 x 148 x 1517 x 1641 x 356 x 1545 No no. Metatarsus Find no. Dextra x 148 x 1479 x 2371 x 728 x 1654 x 1473 x 216 x 1527 x 1435 Sinistra x 1681 x 465 x 1405 x 1399 x 1399 x 1520 x 1430 x 376 x 289 Dex. / Sin. x 148 x 1516 x 1451 x 752 x 870 x 221 x 1403 x 1413 x 1550 x 1473 x 1550 x 513 x 2439 x 1533 x 522 x 763 x 858 x 1032 Square Layer GL Stray find 58.0 Stray find 56.6 106/103 30-40 51.0 110/101 30-40 53.8 93/101 40-50 54.3 107/99 40-50 56.3 106/108 20-30 50.9 107/100 30-40 56.8 107/101 30-40 47.2 94/101 50-60 56.7 From 2001 51.5 103/101 40-50 56.1 109/103 20-30 53.1 107/104 30-40 56.7 105/100 20-30 50.7 Stray find 56.2 107/104 10-20 54.7 106/104 40-50 53.4 94/101 20-30 51.2** Stray find 55.7 105/103 10-20 54.1* 110/100 30-40 56.3 101/101 0-10 56.6 105/108 80-90 55.8 109/103 30-40 55.4 Square Layer GL Bp Stray find 130.3 19.4 105/105 30-40 20.2 From 2001 20.0 106/105 30-40 18.5** 105/107 30-40 21.7 106/103 40-50 19.2 102/100 40-50 20.2 105/109 Ditch 19.5 93/101 20-30 18.1** 105/108 30-40 118.5* 17.7 108/104 20-30 126.8* 108/102 30-40 18.0 106/102 30-40 19.8 106/102 30-40 20.5 105/103 40-50 19.9 107/103 50-60 19.8 106/104 40-50 17.6 107/104 30-40 16.8 Stray find 105/102 40-50 106/98 30-40 107/105 20-30 105/101 30-40 107/104 20-30 106/102 30-40 105/104 40-50 107/98 40-50 106/103 40-50 107/98 40-50 106/101 30-40 From 2001 106/98 50-60 108/104 30-40 106/97 40-50 103/101 0-10 109/99 30-40 ** Juv. SD Bd 11.9 22.2 9.7 21.1 10.4 21.5 23.0 24.7 23.8 21.2 22.4 23.8 23.7 25.8 22.5 24.5 20.8 22.9 22.7 23.1 22.1 21.4 21.9 21.5 ** juv. ** juv. 74 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. Capra hircus Scapula Find no. Dextra x 1503 x 1396 x 1531 Sinistra x 1481 x 604 x 1664 x 1452 x 1483 x 138 Humerus Find no. Dextra x 604 x 647 x 513 x 1579 x 786 x 1469 x 1640 x 1196 x 1488 x 1627 x 1482 x 753 x 786 x 1568 x 356 x 1615 x 458 x 2410 x 148 x 2335 x 141 x 148 From 2001 x 2449 x 166 x 2372 Sinistra x 1663 x 1520 x 490 x 1422 x 1398 x 832 x 527 x 835 x 1459 x 513 x 1560 x 1637 x 148 From 2001 x 139 x 2410 x 2413 x 2346 x 144 x 141 x 148 x 148 x 167 x 148 x 148 x 144 x 2337 x 2410 x 2353 Square Layer 105/103 20-30 106/109 20-30 106/98 50-60 108/103 40-50 106/105 10-20 107/99 40-50 105/109 bottom layer 107/107 20-30 Stray find Square Layer 106/105 10-20 107/100 20-30 106/101 30-40 105/108 50-60 103/101 40-50 107/102 20-30 109/101 20-30 93/101 50-60 106/108 50-60 110/101 20-30 106/103 40-50 105/100 20-30 103/101 40-50 107/108 20-30 101/101 0-10 110/99 40-50 107/97 20-30 From 2001 Stray find From 2001 Stray find Stray find Stray find 109/105 40-50 Stray find 105/110 60-70 107/99 40-50 105/103 40-50 101/100 30-40 105/109 0-10 Stray find 106/96 40-50 108/101 20-30 105/101 30-40 105/99 20-30 106/101 30-40 106/103 30-40 109/102 Trench O Stray find Stray find Stray find From 2001 105/110 70-90 From 2001 Stray find Stray find Stray find Stray find Stray find Stray find Stray find Stray find 103/108 20-30 From 2001 109/97 30-40 HS DHA 151.5 159.8 Bd BT 28.6 26.2 32.7 30.7 34.3 30.8 29.0 27.1 33.5 30.3 30.8 28.9 33.2 31.3 26.0 24.6 28.8 27.6 31.0 29.0 30.7 28.2 31.4* 27.2* 25.5* 29.3 26.2 26.8* 25.8 24.3 22.5 26.8* 25.3* 31.4 30.2 31.4 29.2 29.2 27.3 32.3 30.6 28.0 26.2 32.3* 31.8* 34.2 31.4 33.2* 32.0 33.5 32.0 31.9 29.6 34.9 30.3 29.5 27.2 30.8 28.0* 33.7 31.3 27.8 26.0 31.6 29.7 30.6 28.2 33.8 32.1 29.4 26.3 23.4** 22.3** 29.2 26.8 30.2 27.5 32.4 30.6 28.9* 27.5 32.9 *** 28.9 31.4 28.5 33.1* 31.3 29.5 28.9 26.9 31.2 29.6 34.0 31.6 30.3* 28.8 32.6 30.4 28.6 26.7 24.7 23.5 33.6 31.2 34.5 31.8 29.3 27.1 LD SLC GLP BG 95.7 18.4 30.3 20.6 17.6 29.1 19.8 18.2 31.8 20.2 21.3 37.6 23.7 21.9 36.2 23.5 20.1 34.5 22.5 21.5 39.1 25.0 19.0 32.8 21.5 22.1 36.8 23.7 ** juv. *** Exostosis formation, laterally 75 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. Radius Find no. Dextra x 1629 x 763 x 1328 x 897 x 1499 x 863 x 648 x 1639 x 1641 x 1467 x 1563 x 1046 x 937 x 870 x 1631 x 148 x 148 x 420 x 1550 x 137 x 1514 Sinistra x 1496 x 1045 x 1479 x 1632 x 1485 x 1459 x 1643 x 1521 x 376 x 1459 x 1449 x 1407 x 1544 x 1407 x 1325 x 937 x 588 x 148 x 148 x 144 x 148 Metacarpus Find no. Dextra x 1400 x 1404 x 1530 x 1649 x 1629 x 148 x 1007 x 894 x 1532 x 1432 x 1544 x 1623 x 1461 x 1616 Sinistra x 893 x 1399 x 148 From 2001 x 149 x 2413 x 611 x 1431 x 1492 x 1623 x 1482 Square 106/99 106/97 105/109 107/96 105/105 105/101 100/101 105/107 110/100 Stray find 108/98 110/101 107/97 105/101 Stray find Stray find Stray find 107/101 107/98 Stray find 110/101 105/103 110/99 105/105 109/100 106/108 105/99 109/102 106/98 106/104 105/99 105/103 107/102 107/98 107/102 103/105 107/97 107/105 Stray find Stray find Stray find Stray find Square 106/102 94/101 106/98 105/107 106/99 Stray find 106/99 108/96 105/96 106/103 107/98 108/99 103/105 106/97 108/96 106/102 Stray find Stray find Stray find 105/110 108/105 105/110 109/100 108/99 106/103 Layer Bp 30-40 27.8 40-50 32.7 30-40 32.3 30-40 30.1** 20-30 27.1 40-50 29.3 20-30 27.4 90-100 29.2 30-40 30.3** 29.2** 30-40 28.2 30-40 30.4 50-60 30.4 30-40 28.6 29.6 28.5 31.2 40-50 31.6 40-50 40-50 30-40 29.1 40-50 27.4 30-40 30.9 20-30 28.5 50-60 31.4** 20-30 32.2 27.4 30-40 28.7 40-50 33.0 20-30 30.7 40-50 32.9 30-40 32.3 50-60 30.7** 30-40 20-30 50-60 30-40 32.6 32.2 29.2 29.9 27.3 Layer GL 30-40 104.2 60-70 40-50 50-60 30-40 30-40 40-50 0-70 40-bottom 50-60 30-40 50-60 50-60 30-40 92.9 30-40 98.2 93.5* 97.4* BFp 26.8 31.0 30.8 28.8** 25.6 27.6 26.5 28.7** 28.6** 27.0 28.3 27.5 27.8 30.1 30.5 27.9 27.0 30.2** 30.9 29.9 29.0 30.7 31.3 29.3** 31.0 Bp 22.8 26.6** 21.7** 21.0** 24.8** 24.5** 25.3** 24.9** 23.5** 24.1*) ** 24.6** 25.8** 24.6** 23.6 SD Bd 18.4** 19.0** ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. 70-90 20-30 10-20 30-40 30-40 40-50 21.0 21.7 25.3** 27.0** 23.4** 26.4** 24.0** 25.6** 24.3 SD Bd 15.8 27.0 16.6** 15.4** 13.6** 16.5** 17.1** 15.2** 14.9** 17.4** 16.1** 16.3** 14.0 24.2 15.1 25.6 15.3 16.4 26.5 16.9** 15.3** 16.1** /ad. / ad. / ad. / ad. / ad. /ad. / ad. / ad. / ad. / ad. / ad. /ad. /ad. 76 28.4 29.3 33.2** 27.6 28.1 29.2 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. Dex. / Sin. x 1482 x 1481 x 1483 x 673 x 763 x 1050 Tibia Find no. Dextra x 148 x 522 x 1627 x 1550 x 863 x 1428 x 2353 x 2457 x 665 x 2425 x 134 x 142 x 1418 Sinistra x 1545 x 1612 x 1649 x 149 x 1431 x 148 x 1533 x 590 x 650 x 149 x 1623 Astragalus Find no. Dextra x 1550 x 1672 x 145 x 167 x 1192 x 1550 x 796 x 2341 x 364 x 1533 x 137 x 1443 x 1447 x 420 x 1630 x 1580 x 785 x 145 x 1528 x 1534 x 720 Sinistra x 647 x 145 x 740 x 299 x 895 x 2420 x 228 x 144 x 142 x 1586 x 146 x 1326 x 1501 106/103 40-50 108/103 40-50 107/107 20-30 108/104 20-30 106/97 40-50 99/101 20-30 Square Layer Bd Stray find 26.2 108/104 30-40 25.7 110/101 20-30 23.8 107/98 40-50 23.6 105/101 40-50 25.3 106/109 30-40 25.5 109/97 30-40 24.5 107/94 20-30 22.5 106/96 30-40 23.5 109/98 50-60 23.4 Stray find 26.0 Stray find 24.1 30.6 26.0 25.7 26.6 25.9 25.4* 106/99 30-40 28.4** ** juv. / ad. 105/108 80-90 30.9 106/99 20-30 26.3 105/107 50-60 24.1* Stray find 26.0* 105/110 10-20 25.7 Stray find 24.9 106/98 50-60 24.5 108/99 30-40 25.2 107/97 30-40 22.6 Stray find 24.7 108/99 30-40 23.7 Square Layer 107/98 40-50 97/101 10-20 Stray find Stray find 106/109 20-30 107/98 40-50 103/101 10-20 From 2001 101/100 10-20 106/98 50-60 Stray find 105/110 ? 107/108 30-40 107/101 40-50 106/99 30-40 107/107 10-20 103/101 30-40 Stray find 105/98 20-30 107/103 30-40 107/104 10-20 107/100 20-30 Stray find 108/101 40-50 108/105 30-40 108/96 30-40 From 2001 107/105 30-40 Stray find Stray find 97/101 20-30 Stray find 105/109 30-40 105/108 30-40 GLl GLm 25.8 24.6 28.3 26.5 26.2 24.7 29.9* 28.6* 26.3 25.0 28.7 26.9 28.8 26.6 30.6 28.3 25.8 24.6 29.5 27.9 26.5 25.1 26.8 25.2 26.0 24.3 26.4 25.3 27.1 26.0 29.6 27.8 30.1 28.4 25.8 24.4 26.2 24.8 28.6 27.4 25.1 23.3 27.5 26.2 27.3 25.4 27.7 26.8 30.2 29.7 27.2* 26.2 28.2 26.5 29.2 27.9 26.8 26.2 25.0 23.8 29.7 27.6 27.7* 26.1 28.3 27.4 29.2 28.2 Bd 16.4 17.0 16.0 20.2* 15.7 19.2 18.5 19.6 16.3 17.7 17.0 17.3 16.2 17.9 17.1 18.7 19.3 16.3 16.3 17.5 15.1 17.8 18.0 18.2 19.7 16.8 17.4 17.9 17.0 16.5 17.4 16.7 18.5 18.6 77 2018 Journal of the North Atlantic G. Nyegaard Appendix 10, Continued. No. 37 x 221 x 1520 x 1647 x 535 x 751 x 1469 x 1544 x 513 x 1501 x 1494 Calcaneum Find no. Dextra x 1049 x 1403 x 465 x 1510 x 516 x 522 x 522 x 1509 x 1416 Sinistra x 134 x 1024 x 148 x 1451 x 1007 x 490 Metatarsus Find no. Dextra x 1629 x 1544 x 554 x 1464 x 1539 x 289 x 1049 x 1007 x 1664 x 376 x 167 Sinistra x 1634 x 1612 x 2458 x 1415 x 2371 x 149 x 136 x 1477 x 1041 x 879 x 1045 x 167 Dex. / Sin. x 1011 x 142 x 522 x 289 x 228 x 1399 x 1195 x 873 107/104 20-30 24.1 22.7 15.4 105/103 40-50 24.0 22.8 15.6 110/101 20-30 29.4 27.5 18.8 108/104 10-20 28.0 27.4 18.1 108/100 20-30 26.6 25.4 16.0 107/102 20-30 27.6 26.4 18.0 107/98 50-60 25.9 24.8 16.4 106/101 30-40 27.8 26.1 17.2 105/108 30-40 27.0 25.6 16.8 106/109 40-50 30.9 29.3 19.5 Square Layer GL 106/99 20-30 56.4 106/102 30-40 53.6 108/104 20-30 50.8** ** juv./ad. 106/98 20-30 52.8 106/100 30-40 56.1 108/104 30-40 56.8 108/104 30-40 52.1 105/99 10-20 56.2** ** juv./ad. 108/103 20-30 53.2 Stray find 50.0 97/101 40-50 57.3 Stray find 54.2 106/98 30-40 56.2 106/99 30-40 60.7** **juv. 101/100 30-40 54.9 Square Layer GL Bp SD Bd 106/99 30-40 110.9 18.7 13.0 23.5 107/98 50-60 102.4* 17.3 12.3 Stray find 20.8** 108/98 40-50 20.4** 105/96 0-70 19.2 107/104 30-40 20.2 106/99 20-30 20.7 106/99 30-40 21.2** 107/99 40-50 20.1** 106/104 40-50 18.6 Stray find 21.1** 106/108 40-50 111.9* 13.5 27.8 106/99 20-30 103.9 16.9 11.9 22.3* 105/94 20-30 105.9 18.1 23.8 105/104 40-50 21.5** 15.9** ** juv. ** juv. ** juv. ** juv. ? ** juv.? ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** juv. ** Sub ad. 109/104 30-40 20.4** 13.6** Stray find 19.9** 12.1** Stray find 20.7** 13.6* 105/98 20-30 19.8* ** 106/99 20-30 18.6** 108/97 40-50 21.0** 110/99 40-50 16.7* ** 11.0** Stray find 20.4** Trench F Stray find 108/104 30-40 107/104 30-40 107/105 30-40 106/102 30-40 105/109 20-30 105/101 40-50 25.6 23.9 24.0 24.0 30.3** 23.5 24.2 24.4 Rangifer tarandus Scapula Find no. Square Layer SLC GLP Dextra x 1325 103/105 20-30 32.7 48.5 BG 33.7 78 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. Sinistra x 1475 x 1492 x 1648 Humerus Find no. Dextra x 1639 x 897 x 1041 Sinistra x 1615 x 838 x 487 Radius Find no. Dextra x 895 x 1649 x 897 x 1613 x 1654 x 1509 x 785 Sinistra x 1021 x 1510 x 264 x 1146 x 1145 x 1450 Metacarpus Find no. Dextra x 1325 x 1467 Dex. / Sin. x 1007 x 764 x 621 x 644 x 895 Tibia Find no. Dextra x 1643 x 893 x 1041 x 1195 x 349 x 1548 Sinistra x 1547 x 897 x 665 x 487 x 1457 x 1457 x 1146 x 650 Astragalus Find no. Dextra x 1561 x 894 x 1647 x 1457 x 1640 x 1648 x 1547 x 897 106/98 40-50 40.5 50.0 109/100 30-40 32.3 43.0 107/99 20-30 42.1 Square Layer Bd 105/107 90-100 50.1 107/96 30-40 44.9 106/99 20-30 41.8 110/99 40-50 45.7 107/97 0-10 42.0 107/100 10-20 45.3 Square Layer Bp Bd 108/96 30-40 47.2 105/107 50-60 46.1 107/96 30-40 44.6 105/108 70-80 42.1 105/107 30-40 42.5 105/99 10-20 37.1 103/101 30-40 40.6 110/99 50-60 45.2 106/98 20-30 44.1 106/104 30-40 44.3 107/99 30-40 45.9 107/99 30-40 47.8 106/98 20-30 48.3 Square Layer Bp Bd 103/105 20-30 33.5 Stray find 34.9 106/99 30-40 39.0 103/101 20-30 38.5 105/100 10-20 40.8* 107/97 30-40 37.8 108/96 30-40 41.7 Square Layer Bp Bd 109/102 ? 57.2 108/96 30-40 38.1 106/99 20-30 35.6 105/109 20-30 41.0 102/101 20-30 35.2 107/98 30-40 33.9 107/98 30-40 34.8 107/96 30-40 37.0 106/96 30-40 37.5 107/100 10-20 37.3 108/98 30-40 35.5* 108/98 30-40 38.6 107/99 30-40 38.6 107/97 30-40 42.2 Square Layer GLl GLm Stray find 45.6* 108/96 40-50 45.8 42.3 110/101 20-30 40.6 38.0 108/98 30-40 42.9 40.1 109/101 20-30 43.2 40.5 107/99 20-30 44.2 42.0 107/98 30-40 43.2 41.4 107/96 30-40 41.6 39.2 35.6 30.1 28.7 Bd 30.6* 28.6 25.6 27.8 27.9 27.4 27.8 25.6 79 2018 Journal of the North Atlantic G. Nyegaard No. 37 Appendix 10, Continued. Sinistra x 361 x 1461 x 612 x 245 x 1520 x 1455 x 1049 Metatarsus Find no. x 1547 x 1525 x 1547 x 1416 x 1612 x 1147 x 1145 x 1457 x 1624 x 1510 x 645 x 685 x 1451 101/101 10-20 103/105 50-60 107/100 40-50 107/101 0-10 105/103 40-50 106/102 20-30 106/99 20-30 Square Layer 107/98 30-40 108/98 20-30 107/98 30-40 108/103 20-30 106/99 20-30 107/99 30-40 107/99 30-40 108/98 30-40 109/99 40-50 106/98 20-30 108/97 30-40 108/97 30-40 106/98 30-40 47.5 44.8 45.5 41.2 43.1 43.8 42.8 Bp 44.5 41.9 42.8 39.6 39.9 41.5 40.6 Bd 27.8 27.7 31.8 25.0* 29.0 25.6* 27.2 Canis familiaris 26.0 37.6 37.2 38.0 42.2 38.7 39.9 38.1 37.7 37.3 36.4 36.2 38.3 Measurement numbers Calvarium Find no. x 1453 Mandibula x 1411 x 707 Atlas Find no. x 752 Scapula Find no. x 1454 Humerus Find no. x 1501 x 142 Radius Find no. x 571 Tibia Find no. x 895 x 1467 Felis catus Calvarium Find no. x 1091 Radius Find no. x 1497 Femur Find no. x 1701 Square Layer 105/109 Bottom 94/101 20-30 106/100 50-60 (NB: Canis sp.) Square Layer 107/105 20-30 Square Layer 106/98 30-40 Square Layer 105/108 30-40 Stray find Square Layer 102/100 30-40 Square Layer 108/96 30-40 Stray find Square Layer 106/108 60-70 Square Layer 106/108 40-50 Square Layer 106/108 50-60 No. 18 18.5 mm* No. 13 (L) 18.0 mm* GB 92.4 SLC 23.8 GL 203.6 GL 163.2 GL 94.6 No. 18 a 10.5 mm* No. 13 (B) 7.6 mm* GLP 30.3 Dp 47.5 Bp 16.5 Bp 23.6 No. 20 (L) 13.3 mm* No. 14 18.0 mm* BG 18.2 SD 13.5 13.6 SD 12.1 SD 8.9 12.3 no. 14 9.5 (dex) 9.8 (sin.) No. 20 (B) 15.5 mm* Bd 36.0 34.5* Bd 20.1 Bd 14.9 22.0 no. 18 39.7 Measurements numbers no. 2 79.3 GL 87.6 GL 98.9 no. 9 32.3* SD 4.9 SD 7.2 no. 19 20.7 no. 25 15.5 no. 27 20.9 80