Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 164 The Archaeology of Greenland Greenland was a major destination for the Norse explorers and settlers of the North Atlantic. The colonists settled in 2 main areas on the southern part of the west coast of Greenland, designated as the Eastern Settlement and the Western Settlement. The Eastern settlement (Fig. 1), furthest to the south, was established at the very beginning of the colonization of Greenland around AD 985. Radiocarbon dates indicate that the Western Settlement was established a little later. The Western Settlement (Fig. 2) was eventually abandoned ca. AD 1400. The Norse had completely left Greenland by the late 1400s (Arneborg et al. 2012a). Archaeological surveys and excavations in the Norse Greenland settlements have taken place since the beginning of the 19th century and continue today. Approximately 560 Norse sites have been recorded in the Eastern Settlement and around 75 in the Western Settlement (The National Museum of Greenland, Ancient Monuments Register, Nunniffiit, Greenland). Apart from a few sites on the outer coast, the farms were scattered in the midand inner fjord regions along the coasts, and along rivers and by lakes where the surroundings were suitable for pasture and hayfields, emphasizing the importance placed on animal husbandry of cattle, sheep and goats. Isotopic (δ13C and d15N) studies of diet (Arneborg et al. 2012a) together with the record of the animal bones (e.g., Enghoff 2003, McGovern 1985) show that the subsistence economy did not depend entirely on terrestrial resources. From the initial settlement period the settlers also exploited the rich, accessible marine fauna (Fig. 3). Dependence on sustenance from the sea, especially seals, increased over time (Dugmore et al. 2012, Ogilvie et al. 2009). Not all the farms are assumed to have been occupied at the same time. New research in the Vatnahverfi region in the Eastern Settlement indicates a dynamic pattern with farms being established and abandoned throughout the period of occupation (Madsen 2014). The Norse societies were traditionally socially stratified, with land tenure as a key element of status and wealth. In the The Peopling of the North Atlantic: Isotopic Results from Greenland T. Douglas Price1,* and Jette Arneborg2 Abstract - This discussion of the isotopic analyses of human samples from Greenland begins with a review of the colonization of the island and a description of the sites and the samples that were collected for analysis. In addition, a brief consideration of the geology and bioavailable 87Sr/86Sr is provided. The analysis of the human data from Greenland follows an introduction to the variation present and observable differences between the Eastern and Western Settlements. Specific sites on Greenland are discussed in some detail in terms of the isotopic data that is available. A summary of dietary and mobility estimates is provided. Non-local individuals are identified and in some cases suggestions of place of origin are made. It is important to remember that Greenland was settled later than Iceland and all the Norse graves are from the Christian period, meaning burial in churchyards with few if any grave goods. Viking Settlers of the North Atlantic: An Isotopic Approach Journal of the North Atlantic 1,*Laboratory for Archaeological Chemistry, University of Wisconsin-Madison, Madison, WI 53706: tdprice@wisc.edu. 2The National Museum of Denmark, Frederiksholms Kanal 12, 1220 Copenhagen Denmark: jette.arneborg@natmus.dk. 2018 Special Volume 7:164–185 Figure 1. The Norse Eastern Settlement on Greenland. Figure 2. The Norse Western settlement on Greenland. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 165 late settlement period, a few magnate farms may have controlled most of the land. Farms with churches and, in some cases, also banqueting halls and warehouses (cf. Arneborg 2006) have been recognized as manor farms and the social and economic centers of their regions. In the Eastern Settlement, 16 farms have been recorded with associated churches. Of those, at least 5 churches were established at the time of initial settlement and closed down during the 13th century, indicating a progressive centralization of power in society (Arneborg 2012). Simultaneously, the churches changed scale and role as they shifted from small congregations of local family groups and their servants to serving a larger public, including the inhabitants of the surrounding farms. In the Western Settlement, 3 farms have been recorded with an associated church. The goal of this study is the isotopic investigation of human tooth enamel from individuals buried on Greenland from the period of initial settlement. Our purpose is to learn more about the origins and movement of the individuals who lived on this large, cold island continent. In previous articles in this volume, the methods we used are explained and the general isotopic background for this region is outlined. In this study, we provide the results of the isotopic proveniencing of the human remains from Greenland and document differences between the Eastern and Western settlements. It is clear that many of the first settlers came from Iceland along with their domesticated animals. It is also clear that individuals moved a good bit between the 2 major settlement areas on Greenland as well as returning to Iceland and in at least one case even to Norway. In the following pages, we summarize the baseline isotopic context on Greenland and present the human enamel isotope data. These data are analyzed and discussed on the scale of Greenland, the Eastern and Western Settlements, the sites with burials, and the individual level. Baseline Geology and Bioavailable Isotopes In general terms, the geology of much of Greenland is exceedingly old, some of the most ancient terrain preserved on earth. 87Sr/86Sr values are normally very high in this context. Hoppe et al. (2003) have estimated that Greenland 87Sr/86Sr geological values fall between 0.725 and 0.755. The 2 major settlement areas are located on different geological terrains, with important consequences for the 87Sr/86Sr values at each. The areas of the Eastern and Western Norse settlements on Greenland are dominated by Precambrian rocks constituting the Proterozoic and the Archaean craton (Kalsbeek 1997, Moorbath and Pankhurst 1976). The Western Settlement, near modern-day Nuuk, lies in the Archaean section of the craton. The Eastern settlement is located on Proterozoic rocks, part of the Gardar province in southernmost Greenland, composed of Paleoproterozoic metamorphic intrusive and metamorphic rock sequences. As Price et al. (2015 [this volume]) discussed, 48 samples of archaeological fauna from the Western and Eastern Settlements have been measured for bioavailable 87Sr/86Sr (Fig. 4). Various species have been sampled including cattle, caribou, arctic fox, and ptarmigan (Price et al. 2015 [this volume]:table 5). Hare and the caribou from the Western Settlement show a dramatic degree of variation from ~0.712 to values >0.760, consistent with the known age of the rocks in the region. The majority of the hare values lie between 0.750 and 0.760. The lower values are very likely from animals living near the coast and consuming vegetation impacted by marine rainfall and sea spray (e.g., Chadwick et al. 1999, Vitousek et al. 1999, Whipkey et al. 2000). The Arctic fox, ptarmigan, and cattle do not show the extreme values seen in hare and caribou and probably reflect the best bioavailable 87Sr/86Sr estimate for the Eastern Settlement. Values for the most part range between 0.711 to 0.716. A total of 34 faunal samples from the Eastern Settlement had a mean value of 0.7133 ± 0.0025, with a minimum value of 0.7065 and a maximum value of 0.7212. Clearly, higher strontium isotope values can be expected for local human tooth enamel in the Western Settlement. Figure 3. δ13C values (‰) in human samples. Time progression is left to right (Blue is AMS date, red is archaeologically dated). The midpoint value for those obtaining half their protein from their domestic animals and half from the marine mammals would be δ13C = -16.3‰, whereas those consuming 25% and 75% marine protein would have δ13C = -17.8‰ and -14.9‰, respectively (Arneborg et al. 2012a). Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 166 Oxygen isotope ratios in carbonate should also vary between the Eastern and Western Settlements. Fricke et al. (1995) report large-scale variation in oxygen isotope ratios in modern precipitation on Greenland and across the North Atlantic. Values vary significantly with latitude along the west coast of Greenland. δ18Oen PDB values in non-Inuit human bone carbonate from the Eastern and Western Settlements vary from approximately -8.0‰ to -4.0‰. Average δ18Oen PDB values are approximately -7.5‰ and -5.5‰ in the Western and Eastern Settlements respectively. Greenland Human Isotope Data Our samples for isotopic proveniencing come from both the Eastern and Western Settlements. No pagan graves have been found in the Greenland settlements to this day, and all our human samples are from burials in Christian graveyards. From the Eastern Settlement we have samples of human remains from 5 locations, designated as ruin groups on Greenland (Fig. 1): 1. Tjodhildes church in Qassiarsuk, Brattahlid, in Tunulliarfik fjord (ruin group E29a); 2. Qorlortup Itinnera, ruin group E35 just Figure 4. 87Sr/86Sr values for archaeological fauna from Greenland, ranked by species. Blue = Eastern Settlement, red = Western Settlement. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 167 north of Qassiarsuk; 3. a church close to Igaliku and the Norse Bishop see Gardar in Igaliku fjord (ruin group E48); 4. the church at Innoqqussaq in Igaliku Kujalleq (ruin group E64); and 5. the church at Narsarsuaq in Uunartoq fjord (ruin group E149), presumed to be a nunnery. From the Western Settlement we have Norse human samples from the church at Kilaarsarfik (ruin group W51), also known as Sandnes (Fig. 2). In addition to the Norse burials, 3 Inuit burials have been included in our study for comparison. Samples of faunal remains for the determination of bioavailable isotope ratios were collected from several of the settlements and are discussed below. The sites are also described to provide the context for the human enamel samples used in the isotopic analysis. The isotope data for human tooth enamel from Greenland is presented in Table 1. We have measured strontium isotopes in human tooth enamel from 50 individuals. The average 87Sr/86Sr value for all samples is 0.7125 ± 0.005. These values are highly variable and generally low, varying from 0.7069 to 0.7314, compared to expectations from the faunal remains. These data are graphed by site in Figure 5, with 87Sr/86Sr values shown in ranked order. Human values for local Greenland Norse with a terrestrial diet might be expected to resemble the bioavailable values from the fauna. Several factors have caused these values to be lower than expected. Some of these individuals may be colonists from Iceland or from Norse homelands in the British Isles, Ireland, or Scandinavia, in which case they would not exhibit the high 87Sr/86Sr values observed for Greenland. A diet of marine foods or terrestrial foods from coastal areas with sea spray, heavy rainfall, or seaweed fertilizers would also lower expected human enamel values toward the known value for seawater of 0.7092. It could also be the case that there are unknown areas in the Eastern or Western Settlement with sources for lower 87Sr/86Sr, although this seems unlikely. The unexpectedly low values among the measured individuals will be addressed in more detail below. There are also significant differences among the sites investigated in this study and with the Inuit values that will also be discussed in more detail in the following pages. The mean value for carbonate δ18O for 50 human enamel samples from Greenland is -7.2‰ ± 1.8 with a minimum and maximum of between -11.3‰ and -3.4‰. A kernel density plot of the distribution of values is shown in Figure 6. The primary mode is at around -6.7‰. The secondary mode on the left side of the plot at approximately -9.0‰ suggests a smaller subset that may represent the differences between the Western and Eastern Settlements. The tertiary mode to the right of the plot peaks around -3.5‰. The mean δ18O value for 3 Eastern Settlements is -6.71‰ ± 1.7 and the mean for the Western Settlement at Sandnes is -9.15‰ ± 1.20. This difference is expected given the fact that δ18O values decrease to the north. There are interesting differences in the Greenland human enamel data in terms of sex. Table 2 presents n, mean, standard deviation, minimum, and maximum for males and females in our sample. There are twice as many females in Figure 5. Ranked 87Sr/86Sr values for Norse tooth enamel from archaeological sites in Greenland. The three gray values are from Inuit burials. E = Eastern Settlement, W = Western Settlement. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 168 Table 1. Human enamel from Greenland, sample information and isotope data. DKNM = Danish National Museum. KNK = Greenland National Museum. KAL:= Department of Forensic Medicine, “Antropologisk Laboratorium.” [Table continued on following page.] Lab # Locality/ruin group Tooth DKNM_Id/NKA_Id KAL_Id Sex Age 87Sr/86Sr δ13C‰ δ18O‰ 14C Cal age (1 sigma) 14C Lab_Id F6463 Sandnes (W51) 7- Skeleton XII KAL 930 ? ? 0.7101 -11.9 -9.6 F5652 Sandnes (W51) PM Skeleton X KAL 928 F 20–25 0.7140 -12.3 -8.9 F5653 Sandnes (V51) PM Skeleton XV KAL 932 F 20–25 0.7211 -11.9 -9.7 F5654 Sandnes (W51) Tooth Skeleton XVIIc KAL 936 F 25–30 0.7122 -13.1 -6.6 F5656 Sandnes (W51) PM Skeleton XXII KAL 947 F 30–35 0.7258 -12.0 -9.0 1045 (1030–1116) AAR-5258 F5657 Sandnes (W51) Tooth Skeleton XXXI KAL 959 F 40–45 0.7158 -13.4 -6.5 1301 (1284–1320) AAR-1147 F5658 Sandnes (W51) Tooth Skeleton XXX KAL 960 F 40–45 0.7194 -12.0 -10.1 1301 (1282–1322) AAR-1145 F5659 Sandnes (W51) Tooth Skeleton XXXV KAL 0964 F 25–30 0.7314 -12.5 -9.2 1307 (1290–1328) AAR-1148 F5660 Sandnes (W51) PM Skeleton XI KAL 929 M 35–40 0.7120 -12.6 -8.4 1297 (1271–1317) AAR-1143 F5661 Sandnes (W51) PM Skeleton K1 KAL 986 M 20–25 0.7176 -10.7 -10.5 F3880 Sandnes (W51) Enamel Skeleton X KAL-928 ? 20–25 0.7110 -13.1 -8.8 1408 (1390–1428) AAR-1144 F3881 Sandnes (W51) Enamel Skeleton XVII c KAL-936 ? 25–30 0.7096 -13.6 -5.8 F5662 Narsarsuaq (E149) PM Skelton 4(I) KAL 997 ? ? 0.7124 -11.2 -7.4 1290 (1280–1305) AAR-6147 F5663 Narsarsuaq (E149) PM Skeleton 6(I) KAL 0999 ? 15–20 0.7163 -11.4 -7.0 1290 (1270–1305) AAR-6149 F5664 Narsarsuaq (E149) Incisor Skeleton 3(I) KAL 996 ? 18/20–35 0.7124 -11.4 -6.2 1340–1390 (1320–1405) AAR-6146 F5665 Narsarsuaq (E149) PM Skeleton 7(I) KAL 1000 M 25–30 0.7185 -11.7 -11.3 F5666 Narsarsuaq (E149) LPM Skeleton 10(I) KAL 1001 M 18/20–35 0.7116 -12.8 -6.5 1389 (1312–1414) AAR-1264 F5667 Narsarsuaq (E149) Enamel Skeleton 4(II) KAL 1004 F 18/20–35 0.7116 -12.3 -7.2 F5668 Narsarsuaq (E149) Enamel Skeleton 9(II) KAL 1009 F >35 0.7114 -13.4 -6.6 F5669 Narsarsuaq (E149) LPM Skeleton 11(II) KAL 1011 F 20–25 0.7147 -13.2 -6.6 F5670 Narsarsuaq (E149) Enamel Skeleton b KAL 1134 ? Adult 0.7135 -12.1 -6.5 F5671 Narsarsuaq (E149) Enamel Skeleton 5(II) KAL 1005 F 18/20–35 0.7224 -11.4 -9.9 F3882 Narsarsuaq (E149) Tooth Skeleton 3(I) KAL 996 ? 18/20 –35 0.7134 -13.5 -4.4 F3883 Narsarsuaq (E149) Tooth Skeleton 6(I) KAL 999 ? 15-20 0.7132 -13.7 -6.2 F5672 Tjodhildes Church (E29a) UPM Skeleton 120 KAL 1084 ? ? 0.7092 -13.1 -4.9 F6463 Sandnes (W51) 7- Skeleton XII KAL 930 ? ? 0.7101 -11.9 -9.6 F5652 Sandnes (W51) PM Skeleton X KAL 928 F 20–25 0.7140 -12.3 -8.9 F5653 Sandnes (V51) PM Skeleton XV KAL 932 F 20–25 0.7211 -11.9 -9.7 F5654 Sandnes (W51) Tooth Skeleton XVIIc KAL 936 F 25–30 0.7122 -13.1 -6.6 F5656 Sandnes (W51) PM Skeleton XXII KAL 947 F 30–35 0.7258 -12.0 -9.0 1045 (1030–1116) AAR-5258 F5657 Sandnes (W51) Tooth Skeleton XXXI KAL 959 F 40–45 0.7158 -13.4 -6.5 1301 (1284–1320) AAR-1147 F5658 Sandnes (W51) Tooth Skeleton XXX KAL 960 F 40–45 0.7194 -12.0 -10.1 1301 (1282–1322) AAR-1145 F5659 Sandnes (W51) Tooth Skeleton XXXV KAL 0964 F 25–30 0.7314 -12.5 -9.2 1307 (1290–1328) AAR-1148 F5660 Sandnes (W51) PM Skeleton XI KAL 929 M 35–40 0.7120 -12.6 -8.4 1297 (1271–1317) AAR-1143 F5661 Sandnes (W51) PM Skeleton K1 KAL 986 M 20–25 0.7176 -10.7 -10.5 F3880 Sandnes (W51) Enamel Skeleton X KAL-928 ? 20–25 0.7110 -13.1 -8.8 1408 (1390–1428) AAR-1144 F3881 Sandnes (W51) Enamel Skeleton XVII c KAL-936 ? 25–30 0.7096 -13.6 -5.8 F5662 Narsarsuaq (E149) PM Skelton 4(I) KAL 997 ? ? 0.7124 -11.2 -7.4 1290 (1280–1305) AAR-6147 Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 169 Table 1, continued. Lab # Locality/ruin group Tooth DKNM_Id/NKA_Id KAL_Id Sex Age 87Sr/86Sr δ13C‰ δ18O‰ 14C Cal age (1 sigma) 14C Lab_Id F5663 Narsarsuaq (E149) PM Skeleton 6(I) KAL 0999 ? 15–20 0.7163 -11.4 -7.0 1290 (1270–1305) AAR-6149 F5664 Narsarsuaq (E149) Incisor Skeleton 3(I) KAL 996 ? 18/20–35 0.7124 -11.4 -6.2 1340–1390 (1320–1405) AAR-6146 F5665 Narsarsuaq (E149) PM Skeleton 7(I) KAL 1000 M 25–30 0.7185 -11.7 -11.3 F5666 Narsarsuaq (E149) LPM Skeleton 10(I) KAL 1001 M 18/20–35 0.7116 -12.8 -6.5 1389 (1312–1414) AAR-1264 F5667 Narsarsuaq (E149) Enamel Skeleton 4(II) KAL 1004 F 18/20–35 0.7116 -12.3 -7.2 F5668 Narsarsuaq (E149) Enamel Skeleton 9(II) KAL 1009 F >35 0.7114 -13.4 -6.6 F5669 Narsarsuaq (E149) LPM Skeleton 11(II) KAL 1011 F 20–25 0.7147 -13.2 -6.6 F5670 Narsarsuaq (E149) Enamel Skeleton b KAL 1134 ? Adult 0.7135 -12.1 -6.5 F5671 Narsarsuaq (E149) Enamel Skeleton 5(II) KAL 1005 F 18/20–35 0.7224 -11.4 -9.9 F3882 Narsarsuaq (E149) Tooth Skeleton 3(I) KAL 996 ? 18/20 –35 0.7134 -13.5 -4.4 F3883 Narsarsuaq (E149) Tooth Skeleton 6(I) KAL 999 ? 15-20 0.7132 -13.7 -6.2 F5672 Tjodhildes Church (E29a) UPM Skeleton 120 KAL 1084 ? ? 0.7092 -13.1 -4.9 F5673 Tjodhildes Church (E29a) PM Skeleton F3 KAL 1091 ? 15-20 0.7089 -13.6 -5.8 F5674 Tjodhildes Church (E29a) PM Skeleton 2 KAL 1029 ? 18/20–35 0.7075 -14.3 -8.3 F5675 Tjodhildes Church (E29a) UI Skeleton 41 KAL 1043 M 35–40 0.7090 -13.8 -7.6 1175 (1061–1226) AAR-1569 F5676 Tjodhildes Church (E29a) Molar Skeleton 41 KAL 1043 M 18/20–35 0.7084 F5677 Tjodhildes Church (E29a) UPM Skeleton 66 KAL 1054 F 18/20–35 0.7087 -14.2 -7.7 985 (909–1017) AAR-1571 F5678 Tjodhildes Church (E29a) Enamel Skeleton 86 KAL 1070 F >35 0.7117 -11.4 -9.5 F5679 Tjodhildes Church (E29a) UPM Skeleton 124 KAL 1654 F 20-25 0.7093 -10.8 -3.4 F5680 Tjodhildes Church (E29a) PM Skeleton 125 KAL 1655 ? Adult 0.7074 -14.3 -7.2 F3884 Tjodhildes Church (E29a) Tooth Skeleton 62 KAL 1052 F 18/20–35 0.7107 -14.6 -3.5 F3885 Tjodhildes Church (E29a) Tooth Skeleton 80 KAL 1064 ? ? 0.7089 -13.7 -3.7 F3886 Tjodhildes Church (E29a) Tooth Skeleton 118 KAL 1083 ? ? 0.7111 -11.8 -8.0 F5221 Innoqquasaq (E64) Molar KNK2655#71 M? Ca. 50 0.7088 -14.8 -5.6 F5222 Innoqquasaq (E64) Molar KNK2655#72 F 18–20 år 0.7118 -14.5 -7.0 1051-1152 AAR-12967 F5223 Innoqquasaq (E64) Molar KNK2655#73 m 55+ 0.7089 -15.2 -6.4 967-1067 AAR-12968 F5224 Innoqquasaq (E64) Molar KNK2655#52 F 30 0.7083 -14.4 -6.7 F5225 Innoqquasaq (E64) Molar KNK2655#70 F 18 - 20 0.7079 -15.9 -6.4 F5226 Innoqquasaq (E64) Molar KNK2655#78 M Ca. 40 år 0.7081 -15.0 -7.0 973-1052 AAR-12969 F5227 Innoqquasaq (E64) Molar KNK2655#68 ? 6-7 0.7191 -13.9 -7.9 F5228 Innoqquasaq (E64) Molar KNK2655#69 F 30–36 0.7090 -15.3 -6.2 F3887 Qoornoq- Nuup Kangerlua Tooth House 3/ske. 16 KAL 1441 ? ? 0.7106 -9.7 -8.0 64V1-I-009 F3888 Uunartoq- Uunartoq Fjord Tooth Grave 8 A KAL 1460 ? ? 0.7100 -11.2 -6.9 60V2-IV-001 F3889 Niaqussat 64V2-III-022 Tooth Skeleton 4 KAL 4031 ? ? 0.7106 -10.4 -6.7 F1852 Igaliku E48) Enamel KNK221x11 ? ? 0.7122 F1854 Qorlortoq (E35) Enamel KNK223x14 ? ? 0.7069 F1855 Qorlortoq (E35) Enamel KNK223x15 ? ? 0.7095 Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 170 our sample and they exhibit a much higher average 87Sr/86Sr than the males as well as a higher standard deviation. A bar graph of ranked 87Sr/86Sr values provides a visual look at these differences (Fig. 7). The highest values generally come from female burials at Sandnes in the Western Settlement, suggesting these are local Greenland women. Values below 0.7092 are likely individuals who arrived from Iceland and include both males and females. That boundary incorporates half the males and less than one-third of the females, suggesting that males were more likely to be immigrants to Greenland and often from Iceland. The women may have come in smaller numbers. A plot of all the Greenland human tooth enamel isotope ratios for strontium vs. oxygen shows 2 groups of individuals (Fig. 8). There is a large and varied group between of -10.0‰ and -3.0‰, and 87Sr/86Sr values of 0.706 and ~0.715. A second group is rather linear with δ18O values around -10.0‰ and 87Sr/86Sr values > 0.715. In this graph, the values from the Western Settlement (i.e., the site of Sandnes) are shown in yellow. Five of the 7 values in the second group are individuals from the Western Settlement, and their δ18O is more negative as would be expected. The strontium isotope values also fit with the higher ratios seen in the region of the Western Settlement. Also of interest in this plot are the Sandnes samples not in the second cluster and the Eastern Settlement samples that are. These samples include 2 burials from Narsarsuaq (E149) (skeletons 5(II) and 7(I)) with ratios that fit with the Western Settlement and 7 burials from Sandnes (W51) (skeletons X, XI, XII, XVIIc, XVIIIc and XXXI) with ratios that fit with the Eastern Settlement. In all likelihood these samples represent individuals born in one settlement and buried in the other. The strontium isotope ratios are too high for these individuals to have originated in Iceland. It is possible that some of these individuals were born in Norway or the northern British Isles, but only rarely are human values above 0.718 observed in Viking Age Norway or Britain. It may also be the case for the Western Settlement samples with lower strontium isotope ratios that diets with a predominance of marine foods may have lowered expected enamel values toward the ratio for seawater. The role of marine foods in the diet can be estimated from δ13C values in collagen and enamel carbonate, and both of these data will be considered in due course. Carbon isotope ratios for 50 enamel carbonate samples from Greenland average -13.01‰ ± 1.46, Figure 6. Kernel Density Plot of δ18O values for 50 human enamel samples from Greenland. Values span from -3.07 to -9.85. Figure 7. Bar graph of ranked 87Sr/86Sr values in tooth enamel for males (left) and females from Greenland. Figure 8. Scatterplot of 87Sr/86Sr and δ18O values for all human samples from Greenland. Table 2. Strontium isotope ratios in males and females on Greenland . Sex n 87Sr/86Sr sd min max M 12 0.7109 0.0035 0.7081 0.7185 F 25 0.7133 0.0062 0.7075 0.7314 -90% -92% -94% -96% -98% -100% Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 171 In general, δ13C values for the individuals from the Western Settlement fall in the more positive half of the graph and indicate a more marine diet. Similar differences as those seen in the graph of 87Sr/86Sr and δ18O values also appear in this graph (Figs. 8, 10), with some individuals buried in the Eastern Settlement exhibiting the high strontium isotope ratios and δ13C values comparable with the samples from the Western Settlement, corroborating the likelihood that these individuals were born in the Western Settlement and moved to the Eastern. These individuals include both males and females. Additional information of importance in the consideration of past diet and mobility comes from carbon and nitrogen isotope ratios in bone collagen in the human remains from these sites (Arneborg et al. 2012a, Nelson et al. 2012a). This information provides an indication of the role of marine foods in adult diets and evidence of dietary differences among individuals. Data from 3 of the sites we have investigated on Greenland are summarized in Table 3 (Arneborg et al. 2012c). Several points are to be noted. Individuals from Sandnes in the Western Settlement have relatively positive carbon isotope ratios and higher nitrogen isotope ratios compared to most of the other sites. Narsarsuaq (E149) has a much more positive carbon and a very high nitrogen isotope ratio compared to E29a, Tjodhildes Church, and is more similar to W51 Sandnes. The high nitrogen and more positive carbon isotope values suggest that marine foods were probably predominant at both Sandnes and Narsarsuaq, while terrestrial foods, with a a minimum and maximum of from -10.8‰ to -14.6‰, respectively. A kernel density plot of the distribution of values is shown in Figure 9 and reveals little other than the rather regular distribution of the data with a slight right skew. The single broad mode levels off around -13.0‰ and declines quickly after 14.6‰. A scatter plot of 87Sr/86Sr and δ13C values for all human enamel samples from Greenland is also informative (Fig. 10). Two clusters are identifiable in this plot, a vertical cluster with most of the samples around 87Sr/86Sr value of 0.710 on the yaxis and δ13C ranging between -10‰ and -16‰. A second linear cluster runs horizontally from 87Sr/86Sr Sr values of 0.715 to 0.731 around the -10 to -12 δ13C values. Again, the higher values, above 0.715, are largely from the Western Settlement (red dots). Figure 9. Kernel density plot of δ13C values for all human samples from Greenland. Values span from -10.78‰ to -14.62‰. Figure 10. Scatterplot of 87Sr/86Sr and δ13C values for all human enamel samples from Greenland. Table 3. Carbon and nitrogen isotope ratios in bone collagen from 4 sites in Greenland (Nelson et al. 2012a). Site δ13C‰ δ15N‰ E29a, Tjodhildes Church mean -18.13 5.30 stdev 0.68 6.30 min -18.9 0 max -16.8 12.78 count 9 9 E149, Narsarsuaq mean -15.70 16.41 stdev 0.74 0.97 min -17.25 13.86 max -14.21 18.64 count 24 24 W51, Sandnes mean -16.31 15.08 stdev 0.68 1.21 min -17.6 12.12 max -14.85 17.09 count 34 34 E64, Innoqquasaq mean -18.71 na stdev 10.59 min -26.00 max -16.42 count 5 -96% -97% -98% -99% -100% Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 172 and she had a small church built for prayers at Brattahlid (Magnusson and Pálsson 1965) (Fig. 11). That is why the church was named Tjodhildes church by the archaeologists when first discovered (Meldgaard 1982). All our samples come from the remarkable, small Tjodhildes church. Ten radiocarbon dates from burials in the churchyard indicate the cemetery was used between AD 900 and 1225 (Arneborg et al. 2012b:13, Lynnerup 1998). Today, only the remains of turf wall are preserved; originally, they may have been protective walls surrounding a wooden building. On the inside, the building was ~3.5 m long and ~2 meter wide. There are no traces of an enclosing wall or ditch, but the position of the burials indicates that the churchyard was circular or oval (Fig. 12). The remains of 155 humans were exhumed from the churchyard, all lying on their backs facing east; their arms were placed along the body in an arm position, which dominated from ca. A.D. 1000 to ca.1250 (Kieffer-Olsen 1993:78). The layout of the church and the churchyard supports the early dating. Reused graves support the argument that the graveyard was used for a relatively long period, and we may expect that both new immigrants and later generations of Greenland-born individuals were buried here. The samples from E29a were a mix of male and female and were young adults in age (Table 1). Two of the sampled skeletons have been raespecially plants, were more commonly eaten at Tjodhildes Church. Archaeological Sites and Isotopic Results In this section, the isotopic data from each archaeological site that was sampled is discussed and then compared with one another in the following section. E29a, Tjodhildes Church/Brattahlid/Qassiarsuk The ruins at Qassiarsuk have been identified as Eric the Red’s farm known as Brattahlid in the Sagas (Arneborg 2010, Edwards et al. 2010). The farm is situated on the broad Qassiarsuk plain close to the head of Tunulliarfik fjord. Erik’s farm developed into one of the largest in Norse Greenland, and the Qassiarsuk plain is one of the more settled locales. More than 60 ruins—including dwellings, byres, stables, barns, storehouses, workshops, and enclosures—are scattered over the plain (Arneborg 2006). The site holds 2 churches. A large stone church dated after AD 1250–1300 and built on top of an 11–12th-century church. The other Tjodhildes church lies a short distance from the 2 later churches and is thought to be the first church on the site, built contemporary with the initial settlement. According to Erik’s Saga, Erik the Red’s wife Tjodhild accepted Christianity shortly after arrival on Greenland Figure 11. A reconstruction of the small structure known as Tojdhildes church on Greenland. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 173 Figure 12. The church, churchyard, and burials at Tojdhildes church, Greenland. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 174 which should provide a sense of the range of local bioavailable strontium isotope ratios. At the same time, human consumption of marine foods, and terrestrial food and fodder affected by sea spray and rainfall, would reduce the 87Sr/86Sr value in tooth enamel toward 0.7092. Comparison with the site of E149, Narsarsuaq, also in the Eastern Settlement, may be useful. The samples from this site date from the 13th to the 15th century and do not appear to include any individuals from Iceland (Arneborg et al 2012b:23). The 87Sr/86Sr values for human tooth enamel at Narsarsuaq vary from 0.7114 to 0.7224. All but 1 of the individuals from Tjodhildes Church have strontium ratio values below 0.7114. It is also the case that marine foods in the diet increased over time in Greenland so that we might expect more dampening of high 87Sr/86Sr values at Narsarsuaq, compared to those at Tjodhildes Church. That is indeed the case as indicated by the collagen carbon and nitrogen isotope ratios from these sites (Table 2). Given this information, it is tempting to suggest that the several higher 87Sr/86Sr values at Tjodhildes Church may belong to individuals from Norway or the northern parts of Britain or Ireland. We can examine this hypothesis in terms of the results from oxygen and carbon isotopes in enamel. The mean value δ13C it was -13.24‰ ± 1.32 with with a min–max = -14.62‰ to -10.78‰. Comparison of the enamel carbon isotope ratios from Tjodhildes Church with the rest of Greenland is informative. The mean value for the individuals from Tjodhildes Church is very close to the mean for Greenland and toward the more terrestrial end of the dietary range from marine to terrestrial. Thus, it seems unlikely that marine foods would have a substantial dampening effect on the 87Sr/86Sr values for these persons. Oxygen isotope ratios are also informative with regard to the origins of the individuals found in the churchyard at Tjodhildes Church. δ18O values for all of the individuals from the Eastern Settlement average -6.71‰ ±1.7, as noted earlier. The oxygen isotope ratios for Tjodhildes Church show a wide variation from -9.48‰ to -3.41‰ with a mean of -6.32‰ ± 2.16. For comparison, oxygen isotopes from human tooth enamel for 84 Iceland burials that appear to be local (i.e., 87Sr/86Sr < 0.7092) average -4.94‰ ± 0.87, with a range between -2.23‰ and -6.94‰ (Price and Gestsdóttir 2018 [this volume]). A scatter plot of oxygen vs. strontium isotope ratios from the Tjodhildes churchyard is useful (Fig. 14). Both the low (Burial 125) and high (Burial 118) diocarbon dated. Both samples are of females, one 25–30 years old dated A.D. 909–1017 (±1 sd) (AAR-1571), and the other one was 35–40 years old and dated 1061–1226 (±1 sd) (AAR-1569) (Arneborg et al. 2012b:13) A total of 12 samples were measured for strontium isotopes and 11 for apatite carbon and oxygen. The mean 87Sr/86Sr value for the samples was 0.7092 ± 0.0013 with a min–max = 0.7073 to 0.7117. These values, however, mean little without context and comparison. A bar graph (Fig. 13) of the 87Sr/86Sr values shows an interesting distribution with 2 similarly low values, 7 midlevel values, and 3 high values. The 2 low values fit well with the known range of 87Sr/86Sr from Iceland (Price and Gestsdóttir 2018 [this volume]). The next five values are also likely from Iceland as they fall within the range of baseline values there and below the expected lower limit of 0.7092 for Greenland. Iceland as the homeland for these individuals fits well with the early date for the site from the time of the initial settlement. These individuals could be among the first colonists. There are also 2 samples of cattle from E29, the Brattahlid farm, with values of 0.7064 and 0.7074 that also suggest a place of origin in Iceland. It is important to remember the 87Sr/86Sr value of 0.7092, which is the highest possible value for individuals born and raised in Iceland and approximately the lowest value for those raised on Greenland (shown as a dotted line on Fig. 13). The last 5 individuals from Tjodhildes Church have 87Sr/86Sr values above this limit and were not raised on Iceland. It is important to determine the range of human values on Greenland for locals and nonlocals. Cattle raised in the Eastern Settlement on Greenland have values varyin from 0.7115–0.7160, Figure 13. Bar graph of ranked 87Sr/86Sr values from the burials at E29a, Tjodhildes Church. Red = female, blue = male, gray = unknown. The dashed line marks the value of seawater, 0.7092. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 175 Nevertheless, burials beneath the walls indicate that the church had one or more predecessors of unknown age (Vebæk 1991:25). Radiocarbon dates of excavated skeletons from the churchyard fall within the time period from the middle of the 1200’s to the first decades of the 1400’s (Arneborg et al. 2012b:23). Twelve samples of skeletons from the churchyard at E149 are included in this study. Six samples are from what the excavator C.L. Vebæk (1991) called “Grave field I” Here 2 layers of burials were excavated and the skeletons have later been radiocarbon dated to the 14th century (Arneborg et al. 2012b:23, Lynnerup 1998). Three samples are from “grave field 2” where only the upper layer of burials was excavated, and most probably they are from the same period as the burials in “grave field 1” (Table 1). The 12 samples were measured for strontium, carbon, and oxygen isotope ratios. The mean value for 87Sr/86Sr was 0.7143 ± 0.0033 with a min–max = 0.7114–0.7224. All of the samples have values above 0.7092 and cannot have originated in Iceland. A bar graph of the distribution of rank-ordered values is shown in Figure 15 and documents the substantial variation in 87Sr/86Sr values in these samples. δ18O values average -7.2‰ ± 1.8 and exhibit an extremely wide variation from -4.4‰ to -11.3‰. The wide range of values suggests that some of the individuals buried at this site were non-local. A plot of 87Sr/86Sr vs. δ18O values (Fig. 16) confirms this impression. Two individuals stand out distinctly with the highest 87Sr/86Sr and the most negative δ18O values. It seems very likely that these 2 individuals are originally from the Western Settlement. The third highest 87Sr/86Sr value has an oxygen isotope ratio that fits well with the rest of the samples from Narsarsuaq and probably indicates inclusion among the locals. Oxygen isotope values average strontium isotope ratios have very negative δ18O values. These more negative values should come from more northerly areas of origin and may suggest, for example, that the 2 highest strontium isotope ratios from this churchyard (Burials 86 and 118) were individuals from the Western Settlement. The lower 2 strontium isotope values with more negative oxygen are puzzling, however, and do not fit with baselines in known areas. The strontium isotope ratio points to an Icelandic origin, but the oxygen isotope ratio does not fit. Iceland oxygen isotope ratios measured in human tooth enamel have a mean value of -4.7‰ ± 1.1 and both of these values are more negative than -8.0‰. Interpretation of these 2 low strontium, more negative oxygen ratios is difficult. E149, Narsarsuaq Ruin group E149 at Narsarsuaq in Uunartoq fjord has been identified as the convent Nonne Kloster mentioned by Ívar Bárðarson in his Greenland Description from the later part of the 1300’s (Jónsson 1930:23, Vebæk 1991). The identification is entirely based on the interpretation of the written sources. From an archaeological point of view, the farm at Narsarsuaq does not differ significantly from other farms in the region. This ruin group was one of the larger farms in the Eastern Settlement, with more than 20 structures including the single-room church, residence, byre, stable, barns, and other outhouses. The farm is situated in the southern part of the Eastern Settlement, and the potential for pastoral farming is not the best compared to the central area around Qassiarsuk/Brattahlid and Igaliku/Gardar. A possible pilgrimage route to the warm springs on the nearby island of Uunartoq may have provided an economic basis for the farm. Single-room churches in Greenland are dated to post ca. AD 1250–1300 (Roussell 1941). Figure 15. Bar graph of ranked 87Sr/86Sr values from the burials at E149, Narsarsuaq. Red = female, blue = male, gray = unknown. Figure 14. Scatterplot of Tjodhildes Church strontium and oxygen isotope ratios. Red = female, blue = male, gray = unknown. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 176 -6.7‰ ± 1.7 for 35 samples of human enamel from the Eastern Settlement. There is 1 individual with a less negative δ18O value that might be non-local, although the strontium isotope ratio fits well in the local group. The less negative oxygen value points to a southerly origin, but at the same time is not completely out of the local range. The δ13C values average -12.3‰ ± 0.9, with a narrow degree of variation from -11.2‰ to -13.7‰. This is a narrow span of values suggesting similar diets for the individuals in the cemetery at Narsarsuaq. A plot of 87Sr/86Sr vs. δ13C values (Fig. 17) provides more information. The same individuals with high strontium isotope ratios also have a more positive carbon isotope ratio suggesting more marine foods in the diet and reiterating the likelihood that at least the two highest strontium values are non-local. There are also 2 individuals with 87Sr/86Sr values that are more average and marine carbon isotope ratios that are more positive. It is tempting to propose that the 2 pairs of samples with very similar 87Sr/86Sr vs. δ13C values in the graph may well have belonged to the same households, but we cannot prove such an assertion. E64, Innoqquasaq/Vatnahverfi For the study of the people of the Vatnahverfi region in the Eastern Settlement, we measured strontium, carbon, and oxygen isotopes in the tooth enamel from 9 burials from the site of Innoqquasaq in Igaliku Fjord. The church at E64 is very similar to the other small churches—E29a, E35 and E48— in this study. It is built of stones and turf. The circular churchyard surrounding the church measured ~20 m across. Twelve ruins have been recorded on the site including a dwelling, byre, stable, barn, and other outhouses. Like the churches at Igaliku E48, Qassiarsuk Tjodhildes Church E29a, and E35 in Itinnera, the landnam farm at E64 lost its church rights and social position during the 1200s, presumably to the nearby farm E66, in Igaliku Kujalleq, that developed into one of the largest manors in the Eastern Settlement. When the E64 farm was abandoned is unknown (Arneborg 2012). Archaeological excavations in the churchyard in 2008 and 2010 exposed in situ single graves, several reburials, and a common grave containing 15 individuals (Fig. 18). Arm positions of the dead and radiocarbon dates indicate that the churchyard was established immediately after settlement and was in use until about AD 1200. Three samples are from the common grave. Two samples are from immediately above the common grave, and the remaining 4 are from individual burials. Radiocarbon dates of skeletons in the common grave and of the skeletons just above the grave indicate a date from before AD 1000. The results of the isotopic analyses of the 9 samples from E64, Innoqquasaq, are provided in Table 1) along with other information about the burials. The human samples from E64 have a mean strontium isotope ratio of 0.7101 ± 0.0036, with a minimum value of 0.7079 and a maximum of 0.7191. Comparison of the 9 human 87Sr/86Sr values from E64 with the archaeological fauna from the Eastern Settlement reveals a significant difference. The mean value of the local bioavailable values from the fauna is 0.7132 while the mean value for Figure 17. Scatterplot of 87Sr/86Sr and δ13C values for human enamel from E149, Narsarsuaq. Figure 16. Scatterplot of 87Sr/86Sr and δ18O values for human enamel from E149, Narsarsuaq. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 177 the 9 human samples is 0. 7101. A bar graph of these values in ranked order reveals 2 samples with significantly higher values than the remaining 7 (Fig. 19). It is important to remember that E64 is one of the earliest sites on Greenland, established at the time of initial settlement. The most obvious interpretation of the 7 values below 0.709 is that these individuals came originally from Iceland and were among the first settlers of Greenland. Isotope values below 0.709 are uncommon on Greenland, in Norway, and the older terrains of the northern UK and Ireland. The 2 higher values are also of interest. Burial 68 is a 6–7-year-old child. Burial 72 is an ~14 yearold female. Because of the very high 87Sr/86Sr value (0.7191) observed in Burial 68, it can easily be argued that this child was born on Greenland and the isotope ratio reflects the bioavailable values of the Eastern Settlement. This 87Sr/86Sr value is in fact one of the two highest human values observed anywhere in the Eastern Settlement. This child probably consumed a significant proportion of seafood in its diet, a pattern confirmed by the fact that this individual has the least negative apatite carbon isotope ratio of the nine individuals from E64 (Fig. 20). Apatite carbon isotope ratios in the E64 human sample have a mean of -14.9‰ ± 0.58, with a minimum of -15.9‰ and a maximum of -13.9‰. A plot of 87Sr/86Sr values vs. δ13C for human tooth Figure 19. Bar graph of ranked 87Sr/86Sr values in human tooth enamel from E64, Innoqquasaq. Red = female; blue = male; gray = unknown. Figure 18. The common grave in the churchyard at the site of E64, Innoqquasaq, Greenland. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 178 from Iceland (-4.7‰) and Denmark (-4.3‰) are from local individuals defined by strontium isotope analysis. Human enamel from Norway has an average δ18O value of -4.4‰, and the Faroe Islands have an average of -3.4‰. Carbonate oxygen isotope ratios in the E64 human sample have a mean δ18O of -6.6‰ ± 0.65 with a minimum value of -7.9‰ and a maximum of -5.6‰. Strontium and oxygen isotope ratios are plotted for the E64 individuals in Figure 21. Oxygen isotopes also show a significantly more negative value in Burial 68, again contrasting Greenland vs. Iceland as place of origin. The δ18O value for Burial 72 is within the range for Iceland, but would also fit well in western Norway. The combination of strontium, carbon, and oxygen isotopes, however, in this individual point toward Greenland as the place of origin. The difference in 87Sr/86Sr values between burials 68 and 72 is pronounced, but falls within the range observed for local Greenlanders (Fig. 10). The wide range of 87Sr/86Sr values is likely due to dietary differences, particularly the proportion of seafood in the diet. Arneborg et al. (1999, 2012a) have demonstrated that the role of seafood in human diets increases over time in Greenland. Diets with more marine input would produce 87Sr/86Sr values in bone and enamel that were lower than expected from terrestrial baseline data and closer to 0.7092, the value of seawater and marine foods. E35, Qorlortup Itinnera The site of E35 is situated inland in the Qorlortoq valley north of Qassiarsuk and ~4 km from the coast. The small site contains a small, stone and turf church and churchyard in addition to 6 structures including a dwelling, byre/stable, barn, and workshops. The churchyard was circular and ~17 m in diameter. The layout and design of the church and churchyard, the positioning of the deceased in the graves, and 4 radiocarbon dates of skeleton from the enamel shows that the 2 individuals with higher strontium isotope ratios are also outliers in terms of carbon (Fig. 19). Burial 68 has the least negative carbon isotope ratio of the group, reflecting a more marine diet in early childhood. Burial 72 also has a negative δ13C value, although it does fall within the Iceland range of values. The 7 individuals from Iceland have carbon isotope ratios in tooth enamel from -14.4‰ to -15.89‰, reflecting slightly varying proportions of marine foods in the diet of these individuals. In addition to the enamel apatite carbon isotope ratios, there are 3 bone collagen carbon isotope ratios obtained along with the radiocarbon dates for these samples. Burial 78 had a δ13C value of -19.7‰, Burial 74 was -18.3‰, and Burial 72 had the most positive of the 3 values at -18.1‰, but again within the range of the Iceland values. Thus, Burial 72 has an 87Sr/86Sr value that cannot be from Iceland, and δ13C values that fit with both Greenlanders and Icelanders. In spite of the poorly understood variability in oxygen isotopes, there is useful information that may provide additional information on human mobility in the past. We have measured δ18O in a large number of samples from various places in the North Atlantic. These data are summarized in Table 4. There is a pronounced difference between Greenland and the other countries to the south and east. The average value for all human enamel samples from Greenland is -7.7‰. Average values Figur 21. Strontium vs. oxygen apatite isotope ratios for human burials from Innoqquasaq. Table 4. Means and 1 standard deviation for δ18O in samples of human enamel from various areas in the North Atlantic. Place Local n Average sd Denmark <0.711 71 -4.3 0.7 Norway -- 15 -4.4 1.2 Faroe Islands -- 11 -3.4 0.7 Iceland <0.709 10 -4.7 1.1 Greenland >0.709 35 -7.7 1.9 Figure 20. Scatter plot of 87Sr/86Sr vs. δ13C for the nine burials from E64, Innoqquasaq. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 179 W51, Sandnes/Kilaarsarfik To the north in the Western Settlement, the church at W51 Kilaarsarfik, identified as the Norse site of Sandnes, is the only Western Settlement church included in this study. Radiocarbon dates and artifact finds indicate that the site was settled from the landnam ca. AD 1000 to depopulation ca. 1400 (Arneborg et al. 2012b). The farm is situated on a lush plain at the head of the Ameralik/Ameralla fjords. The settlement includes only 7 recorded ruins, but several houses may have been lost to erosion at the fjord’s edge. The high-status farm at Sandnes was not large compared to Eastern Settlement farms. The church has a Romanesque ground plan indicating an 11th–12th century date. The building was of stone and turf and, like the church at the Eastern Settlement site of Narsarsuaq (E149), it lacked a western end. No predecessors have been recorded, though some rebuilding seems to have been done (Roussell 1936). We may have Christian burial activities on this spot from the beginning of settlement. The churchyard at Sandnes was excavated by Aage Roussell (1936) in the 1930s. Later radiocarbon dating of selected skeletons from the excavations indicate they are from the time period AD 1021–1428 (±1 sd) (Arneborg et al. 2012b:30). Samples from 12 skeletons are included in this study; of those, 6 have been radiocarbon dated to 1275–1428 (±1 sd) (Table 1). A total of 11 teeth were sampled for strontium, oxygen, and carbon isotopes. The average 87Sr/86Sr was 0.717 ± 0.066. A bar graph of these values in ranked order is shown in Figure 22. Although there is a wide range of values in the distribution with a min–max = 0.7101–0.7314, it is unlikely that any of the individuals buried at Sandnes came from Iceland. Four radiocarbon-dated skeletons from the individuals we have sampled date from the later part of the 13th and the 14th centuries AD, suggesting that these persons are a number of generations removed from the first settlers. In addition to the human samples there are 6 samples of fauna from the site that can contribute to our understanding of bioavailable strontium isotope baselines in this area (Table 6). A caribou and 1 hare have the highest ratios we have recorded on Greenland and presumably came from inland or sheltered parts of the region. The remaining animals have values between 0.7111 and 0.7197 that fit well with the majority of the human samples from Sandnes. Three hare have the lowest values for all the hare measured from the Western Settlement (Fig. 5) and presumably came from the coastal zone where sea spray and rainfall added substantial amounts of marine strontium to the vegetation. graves date the church and churchyard to the early settlement period. The farm likely lost its church rights during the 1200s, supposedly to the church at nearby Brattahlid. The date of abandonment for the farm is unknown. Three samples (2 humans KNK 223x14 and KNK 223x15 and 1 cow KNK 223x1) were measured for strontium isotopes and 2 for carbon and oxygen (Table 5). The low strontium isotope ratios of 3 of the individuals point to Iceland as their place of origin, while the third individual has strontium and oxygen isotopes that are more similar to what we expect for the Eastern Settlement of Greenland. There is also 1 cow sample (KNK 223x1) with an 87Sr/86Sr value of 0.7065, indicating that the animal probably came to Greenland from Iceland, like 2 of the humans. E48, Igaliku Moving from Tunulliarfik fjord to Igaliku fjord, at the head of the fjord we find another small inland church, situated at a small farm (ruin group E48), a 15-minute walk from the small community of Igaliku, where formerly the Greenland bishops lived. Igaliku was then the Norse farm known as Gardar (ruin group E47). Typologically, the church is slightly more recent than Tjodhildes church. It is of turf and stone and measures ~6 m x 4.5 m on the outside. Besides the church, the E48 farm includes 11 structures among which are a dwelling, a byre/barn complex, and other outhouses. It is not know when the farm was abandoned. The surrounding churchyard is 4-sided with rounded corners and measures ~19 x 17.5 m on the outside. A small test trench excavated in the churchyard in 2001 revealed a few graves close to the church. Positioning of the dead and radiocarbon dates indicate the cemetery dates to the early 12th century. Only 1 sample from this site was measured for strontium, oxygen, and carbon isotopes. The sample is from grave 3 that cuts older grave number 4. Charcoal of local wood from grave 3 has been radiocarbon dated to AD 1036–1209 (± 1 sd); however, we cannot for certain rule out the possibility that the charcoal is from grave 4. The skeleton in grave 4 is dated 1042-1169 (± 1 sd), with an 87Sr/86Sr value of 0.7122, δ18O of -7.64‰, and δ13C of -14.11‰. These values are typical for individuals born in the Eastern Settlement of Greenland. Table 5. Isotope ratios from human tooth enamel from E35, Qorlortup Itinnera. 87Sr/86Sr δ13C δ18O 0.7069 -13.7 -4.9 0.7095 -14.8 -7.8 0.7067 Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 180 the burials at Sandnes and suggests that these individuals may have been from different parts of the Western Settlement, or at least consumed foods from isotopically different areas. Inuit Samples Three samples from Inuit burials have been included in our investigations, 1 from the Eastern Settlement and 2 from the Western Settlement area. These burials come from Inuit sites, and all date after the Norse abandonment of Greenland and before recolonization in 1721. The 18th-Century Inuit settlement of Niaquusat (Gulløv 1997) is in the Ameralla fjord, in the Western Settlement area on the same location as Norse farm, W48. Qoornoq in Nuuk fjord is another Inuit settlement, also in the Western Settlement area. The skeleton was found on the platform in house 3. The site has been occupied in several periods perhaps already from the 15th century. The youngest ruins on the sites are from the 18th century. House 3 with its contents has not been dated; it is, however, thought to belong to the older settlement period (15th– 16th centuries AD) (Gulløv 1997). The large Inuit settlement in Uunartoq fjord in the Eastern Settlement is The wide range of human values presumably reflects dietary differences among the buried inhabitants of Sandnes, with high 87Sr/86Sr terrestrial diets dampened by the intake of marine foods among the individuals with lower 87Sr/86Sr values. Several of the burials were radiocarbon dated, and collagen carbon and nitrogen isotope ratios were reported along with the dates. These 6 samples averaged -15.83‰ ± 0.65 and 15.4‰ ± 0.25, for carbon and nitrogen, respectively, indicating a relatively high proportion of marine foods in the diet. The 3 highest 87Sr/86Sr human values (Burials Sk. XXXV, Sk. XXII, and Sk. XV) are puzzling because they don’t appear to have been dampened as much by the intake of marine isotope ratios, although their collagen carbon isotope ratios are very similar to the other samples in the group. The apatite carbon isotope ratios from the tooth enamel tend to confirm this interpretation. δ13C values had a mean of -12.26‰ ± 0.85, with a min– max = -10.71‰ to -13.35‰. Sandnes has the most positive δ13Cen of the 4 large sites that were sampled on Greenland, indicating more marine foods in the diet. A scatterplot of δ13C vs. 87Sr/86Sr (Fig. 23) shows the relationship between these 2 ratios. Values for δ18O averaged -9.15 ‰ ± 1.2 with with a min–max = -6.52‰ to -10.46‰. A scatter plot of δ18O vs. 87Sr/86Sr (Fig. 24) emphasizes the wide high degree of variation in both ratios among Figure 23. Scatter plot of 87Sr/86Sr vs. δ13C for the 11 burials from V51, Sandnes. Figure 24. Scatter plot of 87Sr/86Sr vs. δ18O for the 11 samples from V51, Sandnes. Table 6. 87Sr/86Sr values for fauna from V51, Sandnes. Species 87Sr/86Sr value Caribou 0.7611 Cow 0.7151 Hare 0.7175 Hare 0.7111 Hare 0.7197 Hare 0.7579 Figure 22. Bar graph of ranked 87Sr/86Sr values in human tooth enamel from W51, Sandnes. Red = female; blue = male; gray = unknown. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 181 nent in the diet. The effects of this predominantly marine diet can be seen in the 87Sr/86Sr values; all are close to the value of seawater—Niaquusat (0.7106), Qoornoq (0.7106), and Uunartoq (0.7010). The δ18O values are highest at Qoornoq and lowest at Niaquusat, both sites in the Western Settlement. The value for Uunartoq in the Eastern Settlement is closer to the value from Niaquusat; that finding is unexpected and may reflect a different place of birth. In sum, the Inuit data provides a baseline for largely marine-based diets on Greenland. Comparison of the Greenland Sites For comparison, it is useful to look at the isotope ratios from the 4 Greenland sites with large samples (E64, Innoqquasaq, E29a, Tjodhildes church, E149, Narsarsuaq, and W51, Sandnes; Table 8). The sites with 1 or a few samples will be incorporated later in this discussion. The average values of some of the ratios vary substantially among the 4 sites (Table 8). Strontium isotope ratios are high at the Western Settlement of Sandnes and at the later Eastern Settlement site of Narsarsuaq, E149. These values are low at the 2 early sites in the Eastern Settlement, Tjodhildes Church and Innoqquasaq. In general, 87Sr/86Sr values are lower than expected based on bioavailable values in archaeological fauna because of the role very close to the Norse farm Narsarsuaq (E149), also in our study. The sample is from a skeleton collected in grave 8. Among other things, the grave contained a German coin dated to 1705–1743 (Mathiassen 1936). On the basis of the coin the sample should be dated to the 18th century. Traditional Inuit diets were heavily dependent on marine foods, although a number of different terrestrial and marine species were hunted, fished, collected, and gathered (Nelson et al. 2012b, Pars et al. 2001). The most important source of nutrition was the seal, hunted year-round, but whales, seabirds and their eggs, and fish, especially halibut, were also significant components of the diet (Buchardt et al. 2007, Searles 2002). Terrestrial animals included caribou, musk ox in NE Greenland, ptarmigan, and hare. In addition, a variety of plant foods were collected and eaten including grasses, tubers, roots, berries, and seaweed (Pars et al. 2001). Carbon isotope ratios in tooth enamel are similar and relatively high for the range of Norse values from Greenland (Table 7). δ13C values from these individuals clearly reflect a major seafood compo- Table 7. Isotope ratios for Inuit burials from Greenland. Site 87Sr/86Sr 13C 18O Qoornoq 0.710581 -9.73 -8.01 Uunartoq 0.709988 -11.17 -6.90 Niaquusat 0.710567 -10.39 -6.67 Table 8. Means, 1 standard deviation, minimum and maximum isotope values and number of samples for human enamel from four major sites on Greenland. En = enamel, col = collagen. 87Sr/86Sr 13Cen 18Oen 13Ccol 15Ncol Sandnes (W51) mean 0.7174 -12.26 -9.15 -15.78 15.13 stdev 0.0066 0.85 1.20 0.65 0.25 min 0.7101 -13.35 -10.46 -16.42 14.85 max 0.7314 -10.71 -6.52 -14.8 15.4 count 11 11 11 6 5 Innoqquasaq (E64) mean 0.7101 -14.87 -6.61 -18.71 stdev 0.0036 0.58 0.65 10.59 min 0.7079 -15.89 -7.93 -16.42 max 0.7191 -13.88 -5.59 -6.00 count 9 9 9 5 Narsarsuaq (E149) mean 0.7143 -12.33 -7.15 -15.03 17.27 stdev 0.0033 0.93 1.80 0.31 0.43 min 0.7114 -13.67 -11.28 -15.27 16.63 max 0.7224 -11.24 -4.41 -14.61 17.54 count 12 12 12 5 4 Tjodhilde (E29a) mean 0.7092 -13.24 -6.32 -18.5 stdev 0.0013 1.32 2.16 na min 0.7074 -14.62 -9.48 -18.9 max 0.7117 -10.78 -3.41 -18.0 count 12 11 11 2 Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 182 of marine foods in the diet. In addition, the 2 sites with lower 87Sr/86Sr values in the Eastern Settlement have a number of non-local individuals from Iceland and perhaps Norway or northern Britain among the buried individuals who were sampled. Two of the 3 samples (a cow and a human) from E35, Qorlortup Itinnera, in the Eastern Settlement also exhibited strontium isotope ratios consistent with an Icelandic origin, while the third individual appears to be Greenlandic. The single individual from E48, Igaliku, has a very high 87Sr/86Sr in line with origins in Greenland and a low intake of marine foods. There are also obvious differences in Greenland between the Eastern and Western Settlements in the δ18O data in Table 7. The site of Sandnes is the only series of human burials we have analyzed from the Western Settlement and has an average value of -9.2‰. The average for 33 burials from sites in the Eastern Settlement is -6.7‰. The average values among the various sites in the Eastern Settlement are generally similar and vary from -7.2‰ to -6.3‰. It is important to keep in mind that both Tjodhildes Church and Innoqquasaq E64 contain a substantial number of non-local individuals, most from Iceland. δ18O was measured for 2 of the burials from E35, Qorlortup Itinnera, and values of -4.9‰ and -7.8‰ were obtained. The more negative value came from the individual presumed to be from Iceland and the less negative value from the sample thought to be local, as would be expected. The 1 sample from E48, Igaliku, produced a value of -7.6‰, consistent with the higher strontium isotope ratio (0.7122) and suggestive of an origin in the Eastern settlement. The isotope ratios for the Inuit burials are generally similar with only slight differences in the δ18O values, related related at least in 1 case to to differences between the Eastern and Western Settlements. Carbon isotope ratios are available from both the enamel carbonate and bone collagen and provide complementary information. Both sets of values show a rather narrow spread. δ13Cen values vary from -15.9‰ to -10.7‰ for 43 Norse samples from all sites, with most values clustered between -13‰ and 14‰, and a smaller group between -11‰ and -12.5‰ (Fig. 25). Values for δ13Ccol for 23 Norse bone samples had an average value of -16.9 ± 1.62‰, with a minimum of -19.7‰ and a maximum of -14.6‰. The ratios from enamel and collagen are not directly comparable, but a plot of the values is revealing (Kellner and Schoeninger 2007; Fig. 26). The Kellner model uses regression lines to distinguish diets dominated by C3 protein vs. those with marine protein (and/or C4 plant species). Samples fall on or between 2 two regression lines depending on the proportion of C3 or marine foods in the diet. For both protein types, the diets of primarily marine carbohydrate and lipids fall at the upper end of the line and those with C3 carbohydrate and lipid fall at the lower end of the line. Two clusters of samples are clearly shown in the plot: a group of 5 with δ13Ccol values of -18‰ or less and δ13Cen values of -14‰ or less. These values point to a diet based largely on terrestrial foods without significant marine intake. The second group of 11 samples has with δ13Ccol values of -17‰ or more positive and δ13Cen values of -13.5‰ or more positive. A plot of these values (Fig. 26) on a Kellner diagram (Kellner et al. 2007) provides an indication of the proportion of marine 0.7092 in the diet as well as some reflection of place of origin. Four of the 5 hollow circles in the diagram are samples with 87Sr/86Sr values indicative of an origin in Figure 25. Histogram of δ13Cen for 43 Norse samples from Greenland. Figure 26. Scatterplot of δ13Cen vs. δ13Ccol for 16 Norse samples from Greenland on a Kellner diagram (Kellner and Schoeninger 2007). The hollow circles mark samples with 87Sr/86Sr values indicative of an origin in Iceland. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 183 Iceland. The distinction of 2 diets in the graph is intriguing and clearly marks the predominance of terrestrial diets on Iceland and marine diets on Greenland. Conclusions The isotopic analysis of human remains from Greenland has revealed a substantial amount of information about diet and mobility among the inhabitants of the Eastern and Western Settlements. This information is available on several levels, from comparisons of the population of Greenland with the rest of the North Atlantic, from comparison of the Eastern and Western Settlements, and comparison of the households or communities represented by the various sites. In addition, the isotopic data also provides information at the individual level and makes it possible to discuss a person’s diet and movement in the past in some detail. Greenland, because of the very old bedrock of which it is composed, has generally very high 87Sr/86Sr values. At the same time, the consumption of marine foods by Norse and Inuit peoples has dampened these high ratios below what would be expected from the bioavailable values of terrestrial species. Nevertheless, human enamel 87Sr/86Sr values for individuals born in Greenland lie above 0.7092 and can be much higher. Oxygen isotopes in Greenland are also generally more negative than elsewhere in the Norse Atlantic. Values from 49 samples on Greenland average -7.23‰ ± 1.78‰ and vary from -11.28‰ to -3.42‰ and are distinct from Iceland, Norway, and the northern British Isles and Ireland. A plot of δ18O vs. 87Sr/86Sr shows this pattern for Greenland, Iceland, and Norway (Fig. 28). Comparable data from the British Isles and Ireland are not yet available. Oxygen isotope ratios are useful for distinguishing Greenland from Iceland and Norway, although the latter 2 groups overlap substantially. In part, because of the specific sites that have been excavated and sampled for this study of Greenland, there is a high proportion of Icelanders in the population, individuals who were among the first settlers, especially at the sites of E29a, Tjodhildes Church, and E64, Innoqquasaq. The newcomers from Iceland included cows and probably other domestic species as well. There are also important useful isotopic differences between the Eastern and Western Settlements. The geology of the Western Settlement is somewhat older than that of the Eastern Settlement, which results in significantly higher 87Sr/86Sr values. Human diet in the Western Settlement was more marine oriented. The last humans from the Eastern Settlement were as marine-based as the last from the Western Settlement (Arneborg et al. 2012c). This marine component does dampen 87Sr/86Sr values to some extent, but recognizable differences remain. Oxygen isotope ratios are also more negative at the more northerly Western Settlement. Because of these differences it is possible to discern individuals who moved from the Eastern to the Western Settlement and, in one or two cases, individuals who went the other way. Isotopic analysis at the individual level is of particular interest. We can see for example at E64, Innoqquasaq, that 2 young individuals were likely born locally, while their parents and relatives in the cemetery came to Greenland from Iceland at this early settlement. There were also a substantial proportion of Iceland natives at the early sites E29a, Tjodhildes Church, and E35, Qorlortup Itinnera. There was surprising variation in the strontium and oxygen isotope ratios at W51 Sandnes in the Western Settlement. The data suggest either varied diets among the inhabitants or multiple places of origin within the Western Settlement, or both. Figure 27. Scatterplot of δ13Ccol vs. δ15Ncol for 16 Norse samples from Greenland. Figure 28. Scatterplot of δ18O vs. 87Sr/86Sr for human and faunal samples from Greenland, Iceland, and Norway. The diagonal line is an approximation of the separation of these samples. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 184 The 3 Inuit individuals analyzed as part of this project confirm a heavily marine diet, as is reported ethnographically, and the significant dampening of terrestrial bioavailable 87Sr/86Sr to values very close to 0.7092. These data reaffirm the picture of these people as successful marine foragers, known to have expanded in numbers and territory as the Norse abandoned Greenland in the face of climatic hardship. Literature Cited Arneborg, J. 2006. Saga Trails. Vintervår, Linde Tryk, Denmark. Arneborg, J. 2010. Brattahlids beliggenhed. Grønland 4:320–328. Arneborg, J. 2012. Churches, Christianity and Magnate Farmers in the Norse Eastern Settlement. Pp. 167–170, In H.C. Gulløv, P.A. Toft, C.P. Hansgaard (Eds.), Northern Worlds: Challenges and Solutions. Report from workshop 2 at the Nationalmuseum, 1 November 2011. Copenhagen, Denmark. 373 pp. Arneborg, J., J. Heinemeier, N. Lynnerup, H.L. Nielsen, N. Rud, and A.E. Sveinbjörnsdottir. 1999. Change of diet of the Greenland Vikings determined from stable carbon isotope analysis and 14C dating of their bones. Radiocarbon 41:157–168. Arneborg, J., J. Heinemeier and N. Lynnerup. (Eds.). 2012a. Greenland Isotope Project: Diet in Norse Greenland AD 1000–AD 1450. Journal of the North Atlantic. Special Volume 3. Arneborg, J., N. Lynnerup, J. Heinemeier, J. Møhl, N. Rud and A., Sveinbjörnsdóttir. 2012b. Norse Dietary Economy ca. A.D. 980–ca. A.D. 1450: Introduction. Pp. 1–39, In J. Arneborg, J. Heinemeier, and Niels Lynnerup (Eds.). Journal of the North Atlantic. Special Volume 3:1–39. Arneborg, J., N. Lynnerup and J. Heinemeier 2012c. Human Diet and Subsistence Patterns in Norse Greenland AD c.980–AD c.1450: Archaeological Interpretations. Pp. 119–133, In J. Arneborg, J. Heinemeier, and N. Lynnerup (Eds). Journal of the North Atlantic. Special Volume 3:119–133. Chadwick, R.A., D.I. Jackson, R.P. Barnes, G.S. Kimbell, H. Johnson, R.C. Chiverrell, G.S.P. Thomas, N.S. Jones, N.J. Riley, E.A. Pickett, B. Young, D.W. Holliday, D.F. Ball, S.G. Molyneux, D. Long, G.M. Power, and D.H. Roberts. 2001. Geology of the Isle of Man and its offshore area. British Geological Survey, Nottingham. Dugmore, A.J., T.H. McGovern, O. Vésteinsson, J. Arneborg, R. Streeter, and C. Keller. 2012. Cultural adaptation, compounding vulnerabilities and conjunctures in Norse Greenland. Proceedings of the National Academy of Science 109:3658–3663 Edwards, K.J., J.E. Schofield, and J. Arneborg. 2010. Was Erik the Red’s Brattahlið Located at Qinngua? A Dissenting View. Viking and Medieval Scandinavia 6:83–99. 10.1484/J.VMS.1.102137. Enghoff, I.B. 2003. Hunting, fishing and animal husbandry at The Farm Beneath The Sand, Western Settlement. Meddelelser om Grønland, Man & Society 28. Copenhagen, Denmark. 104 pp. Fricke, H. C., J.R. O’Neil, and N. Lynnerup. 1995. Oxygen isotope composition of human tooth enamel from medieval Greenland: Linking climate and society. Geology 23:869–872. Froehle, A.W., C.M. Kellner, and M.J. Schoeninger. 2012. Multivariate carbon and nitrogen stable isotope model for the reconstruction of prehistoric human diet. American Journal of Physical Anthropology 147:352–69. Gulløv, H.C. 1997. From Middle Ages to Colonial Times. Archaeological and ethnohistorical studies of the Thule Culture in South West Greenland 1300–1800 AD. Meddelelser om Grønland, Man & Society 23. Copenhagen, Denmark. 501 pp. Hoppe, K.A., P.L.Koch, and T.T. Furutani. 2003. Assessing the Preservation of Biogenic Strontium in Fossil Bones and Tooth Enamel. International Journal of Osteoarchaeology 13:20–28. Johansen, O.S., S. Gulliksen, and R. Nydal. 1986. 13C and diet: Analysis of Norwegian human skeletons. Radiocarbon 28:754–761. Jónsson, F. 1930. Det Gamle Grønlands Beskrivelse af Ívar Bárðarson. Glydendal, Copenhagen, Denmark 75 pp. Kalsbeek, F. 1997. Age determinations of Precambrian rocks from Greenland: Past and present. Geology of Greenland Survey Bulletin 176:55–59. Kellner, C.M., and M.A. Schoeninger. 2007. A simple carbon-isotope model for reconstructing prehistoric human diet. American Journal of Physical Anthropology 133:1112–1127. Kieffer-Olsen, J. 1993. Grav og gravskik i det middelalderlige Danmark. Aarhus Universitet, Middelalderarkæologisk Nyhedsbrev. Århus, Denmark. 212 pp. Lynnerup, N. 1998. The Greenland Norse: A Biologicalanthropological Study. Meddelelser om Grønland, Man & Society 24. Copenhagen, Denmark. Madsen, C.K. 2014. Pastoral Settlement, Farming, and Hierarchy in Norse Vatnahverfi, South Greenland. Ph.D. dissertation, University of. Copenhagen, Denmark. Magnusson, M., and H. Pálsson. 1965. The Vinland Sagas. Penguin Books, New York City, USA. Mathiassen, T. 1936. The Eskimo Archaeology of Julianehaab District. Meddelelser om Grønland 118(1). McGovern, T.H. 1985. Contributions to the paleoeconomy of Norse Greenland. Acta Archaeologica 54-1983:73– 122. Meldgaard, J. 1982. Tjodhildes Kirke: Den første fundberetning. Grønland 5-6-7:151–162. Moorbath, S., and R.J. Pankhurst. 1976. Further rubidium- strontium age and isotope evidence for the nature of the late Archaean plutonic event in West Greenland. Nature 262:124–126. Journal of the North Atlantic T.D. Price and J. Arneborg 2018 Special Volume 7 185 Nelson, E., J. Heinemeier, N. Lynnerup, J. Arneborg, and Á.E. Sveinbjörnsdóttir. 2012a. An isotopic analysis of the diet of the Greenland Norse. Journal of the North Atlantic Special Volume 3:93–118. Nelson, D.E., N. Lynnerup and J. Arneborg 2012b. A First Isotopic Dietary Study of the Greenlandic Thule Culture. Pp. 51–64, In Arneborg, J., J. Heinemeier, and N. Lynnerup (Eds.). Greenland Isotope Project: Diet in Norse Greenland AD 1000–1450. Journal of the North Atlantic. Special Volume 3:51–64. Ogilvie, A.E.J., J.M. Woollett, K. Smiarowski, J. Arneborg, S. Troelstra, A. Kuijpers, A. Pálsdóttir, and T.H. McGovern. 2009. Seals and sea ice in Medieval Greenland. Journal of the North Atlantic. Volume 2:60–80. Pars, T., M. Osler, and P. Bjerregaard. 2001. Contemporary use of traditional and imported food among Greenland Inuit. Arctic 54:22–31. Price, T.D., K.M. Frei, and E. Naumann. 2015. Isotopic Baselines in the North Atlantic Region. Journal of the North Atlantic Special Volume 7:103–136. Price, T.D., and H. Gestsdóttir. 2018. The peopling of the North Atlantic: Isotopic results from Iceland. Journal of the North Atlantic Special Volume 7:146–163. Rousssell, A. 1936. Sandnes and the Neighbouring Farms. Meddelelser om Grønland 88 (2) Roussell, A. 1941. Farms and Churches in the Mediaeval Norse Settlements of Greenland. Meddelelser om Grønland 89(1). Kommissionen for Videnskabelige Undersøgelser i Grønland. Copenhagen, Denmark. Searles, E. 2002. Food and the Making of Modern Inuit Identities. Food and Foodways: History and Culture of Human Nourishment 10:55–78. Vebæk, C.L. 1991. The Church Topography of the Eastern Settlement and the Excavation of the Benedictine Convent at Narsarsuaq in the Uunartoq Fjord. Meddelelser om Grønland, Man & Society 14. Copenhagen, Denmark. Vitousek, P.M., M.J. Kennedy, L.A. Derry, and O.A. Chadwick. 1999. Weathering versus atmospheric sources of strontium in ecosystems of young volcanic soils. Oecologia 121:255–259. Whipkey, C.E., R.C. Capo, O.A. Chadwick, and B.W. Stewart. 2000. The importance of sea spray to the cation budget of a coastal Hawaiian soil: A strontium isotope approach. Chemical Geology 168:37–48.