Increasing Mobility at the Neolithic / Bronze Age Transition-sulphur isotope evidence from Öland , Sweden

1. Archaeological Research Laboratory, Stockholm University, SE-106 91 Stockholm, Sweden. *Corresponding author: gerik@arklab.su.se 2. Department of Geological Sciences, Stockholm University, SE-106 91 Stockholm, Sweden 3. Department of Archaeology, University of Durham, Durham, DH1 3L, United Kingdom 4. Stiftelsen Kulturmiljövård, Stora Gatan 41, SE-722 12 Västerås, Sweden 5. Department of Archaeology and Social Anthropology, Breiviklia, University of Tromsø, NO-9037 Tromsø, Norway


Introduction
Mobility and migration are of fundamental importance in human prehistory and have, as such, been debated ever since archaeology became an academic discipline.The discussion has focused mainly on the power of migration to explain cultural evolution and social development, and on the implications of mobility and sedentism.Clearly, there have been periods during prehistory when migration and changes in mobility patterns were more intense than during other periods, often in connection with the introduction of new crops, creatures or crafts.Scandinavia during the Neolithic and Bronze Age is one such instance, and the focus of the present article.
Mobility has previously been studied using carbon and nitrogen isotopes (e.g.Hakenbeck et al. 2010;Sealy2006), strontium (e.g.Price et al. 2001;Knudson and Buikstra 2007;Nehlich et al. 2009), or sulphur (e.g.Linderholm et al. 2008;Vika 2009;Oelze et al. 2012;Nehlich et al. 2012;Jay et al. 2013).However, the intricate issue of separating mobility from dietary changes has rarely been addressed (although see e.g.Knudson et al. 2010).In this article, we therefore set out to do this, studying Neolithic and Bronze Age people on the island of Öland in the Baltic Sea.In order to study individual mobility, we have focused on establishing intra-individual δ 34 S variation by analysing, where possible, both tooth and bone elements from each individual, enabling the detection of residential changes during a lifetime.This is particularly challenging in aquatic environments, with individuals consuming various mixtures of terrestrial and marine foods, because the terrestrial δ 34 S signal is masked by marine δ 34 S influence.Here, we suggest that by modelling the δ 34 S of the terrestrial component of human diet, it is possible to identify non-local origin and residential mobility for individuals consuming various mixtures of terrestrial/marine protein.In this study we accordingly make use of intra-individual data to distinguish between mobility and dietary change, by combining carbon, nitrogen and sulphur stable isotope data.

Archaeological Background
Figure 1: Geological map over Öland with the location of the archaeological sites analysed, as well as Swedish provinces and sites mentioned in the text.
In the region surrounding the brackish Baltic Sea, the water itself of course served as an excellent medium for mobility, but the Baltic was also of immense importance for subsistence, offering a wealth of resources for the provision of food, clothing, artefact production and fuel, to name just a few examples.The material analysed in this study originates from Öland, a c. 140km long island in the Baltic Sea (Figure 1).It is a narrow island, less than 20km across, in close proximity to mainland Sweden.The bedrock comprises primarily sedimentary rocks such as Ordovician limestone, with some Cambrian and Ordovician shales.Its current shape and relative closeness to the mainland have only changed marginally since the Mesolithic (Svensson 2001; for a chronological division of archaeological periods in southern Scandinavia, see Table 1), and while the natural boundaries of the island itself delimit the area of investigation, the proximity to the mainland still promotes contacts and mobility of people as well as animals.Furthermore, the calcareous soils on the island provide favourable preservation conditions for skeletal material.Öland is one of the few places in eastern Sweden where megalithic tombs occur.The erection of megalithic tombs such as dolmens and passage graves, clearly associated with the Funnel Beaker culture (the TRB), took place over vast areas of northern Europe during the Early and Middle Neolithic, around the middle of the fourth millennium BC (Midgley 2008).In Sweden, about 525 dolmens and passage graves are known (Sjögren and Price 2013), at least 255 of which are concentrated in the Falbygden area (Figure 1), located in the interior of the Swedish mainland, where a limited area of young sedimentary bedrock, primarily limestone, is surrounded by much older, Precambrian, igneous rock.The megaliths outside of the Falbygden area occur mainly along the coasts of the provinces of Bohuslän, Halland and Scania, with a few notable exceptions: the Alvastra dolmen in the province of Östergötland (located only a few kilometres from the Alvastra pile dwelling), a dolmen on the island of Gotland, and three passage graves and a dolmen in Resmo parish on Öland (Figure 1).
The Middle Neolithic in Southern Scandinavia (c.3300-2300 BC) is characterised by the presence of archaeological remains associated with three different, partly coeval, material cultures.The Funnel Beaker Culture is the first farming culture in this region.It is followed in the archaeological record by the Battle Axe Culture (a regional version of the Corded Ware Culture), traditionally perceived as pastoralists.Chronologically partly overlapping with these two cultures is the Pitted Ware Culture (PWC), mainly found at coastal sites and perceived as marine hunter-gatherers.A primary focus for discussion in Scandinavian research on the Middle Neolithic concerns whether these differences in material culture can be attributed to different groups of people, or if the differences mainly reflect different activities of the same group (Lidén and Eriksson 2007 and references cited therein).
The overall importance of food and diet -not only for survival, but also for the construction of identity and culture -makes it crucial for archaeological understanding of cultural differences and change.In a previous study, we therefore used δ 13 C and δ 15 N data in order to reconstruct dietary patterns at several sites on Öland, chronologically including primarily the Neolithic and Early Bronze Age (Eriksson et al. 2008).In our study it became evident that in fact there were differences in diet between the Funnel Beaker Culture and the Pitted Thus, while the Middle Neolithic Torsborg diet was characterised by various mixtures of terrestrial and marine resources with substantial intra-individual differences, the Late Neolithic/Early Bronze Age diet was homogeneous and solely dependent on terrestrial resources.
The notion of two separate Middle Neolithic groups of people in this region -rather than different endeavours by only one group -is further supported by DNA studies.The genetic analysis of human remains associated with the Funnel Beaker Culture on the Swedish mainland on the one hand, and with the Pitted Ware Culture on Gotland and Öland on the other, suggests that these two material cultures represent two different genetic populations (Linderholm 2008;Malmström et al. 2009;2010;Skoglund et al. 2012;2014).
Following the cultural diversity of the Middle Neolithic and the more homogeneous Late Neolithic, we observe a boom in the archaeological record at the onset of the Bronze Age, including e.g.artefacts, settlement patterns and burial customs.These changes coincide chronologically with changes in δ 13 C and δ 15 N values observed on Öland.Our question is whether this can be explained in terms of increased contacts and interaction between different geographical regions and/or cultural groups.Here, we investigate the level of mobility and contacts during the Neolithic and Early Bronze Age by means of stable sulphur isotope analysis.

Stable sulphur isotopes
Sulphur is incorporated into vertebrates through their diet.Experimental data has shown that it is mainly the protein portion of the diet that is reflected in collagen isotopic data (Ambrose and Norr 1993).Sulphur in collagen is present in only two amino acids, methionine and cysteine.Methionine is an essential amino-acid, which implies that it is derived directly from ingested protein, whereas the cysteine is non-essential, and synthesised either from the diet or from methionine (Bohinski 1979).There are four naturally occurring sources for sulphur: organic matter, minerals in rocks and soils, sea water, and atmospheric deposition of sulphuric gases.The δ 34 S value in plants generally primarily mirrors that of its geological surroundings (Brady and Weil 1999).The terrestrial sulphur isotopic signature thus varies depending on the geological setting, and terrestrial δ 34 S values are far more varied than in marine environments.In sedimentary rocks, the δ 34 S values range from -40 to +40‰.European granitic rocks display δ 34 S values between -4 and +9‰, mafic rocks have δ 34 S values close to 0‰, and metamorphic rocks exhibit δ 34 S values ranging from -20 to +20‰ (Krouse 1980;Faure and Mensing 2005).However, the bioavailable sulphur displays ranges that are levelled out compared to the ranges in bedrock.The δ 34 S values for the oceans, by contrast, are rather uniform, averaging +21‰, with marine vegetation having δ 34 S values between +17‰ and +21‰ (Peterson and Fry1987).Due to the so-called sea spray effect (Kusakabe et al. 1976;Wadleigh et al. 1994), sulphur isotopes in plants growing close to the shore might be affected by marine sulphur to some extent.Freshwater systems are much more dispersed, the δ 34 S values ranging from -22 to +20‰, as a result of the reduction of sulphate ions (SO 4¯) to hydrogen sulphide (H 2 S) (Krouse 1980;Faure and Mensing 2005).
The isotopic fractionation between food and consumer is relatively small (-1‰ to +2‰), which means that the δ 34 S value in bone or dentine collagen reflects the sulphur isotopic composition of the diet which, in turn, reflects the geology/locality where the food sources originated (Peterson et al. 1985;Bol and Pflieger 2002;Sharp et al. 2003;Richards et al. 2003;Fraser et al. 2006;Buchardt et al. 2007

Analysed material
The sulphur isotope dataset presented here derive from human and faunal skeletal remains from three sites on Öland: Resmo, Köpingsvik and Torsborg (Figure 1).All analysed human subjects have been radiocarbon dated, spanning from the Middle Neolithic to the Early Bronze Age (c.3500-1000 cal BC), whereas the faunal material is only partly dated, covering the same time period, but also extending the date range into modern times (Eriksson et al. 2008; the present study) (Table 2 and Table 3 -see end).Only samples previously analysed for δ 13 C and δ 15 N that fulfilled the collagen quality criteria with regard to collagen yield (van Klinken 1999 C/N ratio (DeNiro 1985) and C-and N-concentrations (Ambrose 1990) were selected for δ 34 S analysis.In addition, we present radiocarbon dates for Resmo subject 31 -confirming its previously suggested Neolithic date -and for one Middle Neolithic and one Bronze Age cattle specimen from Köpingsvik (see Table 2 and Table 3).
Stable sulphur isotope data, along with other isotope data, for the Resmo individuals have been presented in a previous paper about the Neolithization on Öland (Eriksson et al. 2013).
With regard to δ 34 S analysis, a number of samples have since been rerun, with some additions, and with a more rigorous application of quality criteria.These measures have resulted in a smaller but hopefully more reliable dataset for Resmo, although as there are now fewer samples for some individuals, the interpretation regarding mobility has accordingly changed in some instances (Table 4).
The faunal samples, including both wild and domestic terrestrial species, as well as marine mammals, were included to serve as a necessary baseline for the interpretation of human sulphur isotope data.Their respective ranges of sulphur isotope values enable prediction of the local terrestrial range, reflecting the bioavailable δ 34 S composition of the island, and also allow modelling of the δ 34 S of the terrestrial component of the human diet (see Section 5 for details).

Collagen extraction and sulphur isotope analysis
Collagen, the main protein component of bone as well as dentine, was extracted from the skeletal material in a laboratory dedicated to bone chemical analysis.The skeletal specimens were cleaned using deionised water and the outer surface was removed prior to sampling.Bone or dentine powder was obtained from each sample using a dentist's drill.
Tooth samples were taken from the crown and cervix of the tooth, unless otherwise stated.
Collagen was subsequently extracted following a modified Longin method (Brown et al. 1988).In short, this method includes the following steps: samples were demineralised for approximately 48 h in 0.25 M HCl.Inorganic materials were subsequently removed by filtration and the remaining organic material was rinsed in deionised water and then gelatinised in 0.01 M HCl at 58ºC for c. 16 h.The solution was filtered in a 30 kDa ultrafilter in order to eliminate fragmented collagen peptides, as well as other contaminants such as some humic substances.The residual >30 kDa fraction was frozen to -80ºC and freeze dried overnight.
The sulphur isotope analyses were performed using the EA-IRMS technique either at Iso-

Quality
Of the analysed material from Resmo, Köpingsvik and Torsborg, 53 human and 18 faunal bone and dentine samples fall outside the acceptable ranges for δ 34 S analysis with regard to %S, C/S and N/S (cf.Nehlich and Richards 2009).These samples are reported in Table 5 (see end), together with human samples from three additional Neolithic and Bronze Age sites on Öland (previously analysed with regard to δ 13 C and δ 15 N, see Eriksson et al. 2008), where no samples met the quality requirements.There is no correlation between %S and δ 34 S and accordingly no indications that the δ 34 S values have been affected by post-depositional contamination (cf.Kinaston et al. 2013).
In total, 70 human and 38 faunal samples fulfil the quality criteria with regard to %S, C/S and N/S, as suggested by Nehlich and Richards (2009).These include seven human and four faunal samples (highlighted by an asterisk in Table 2 and Table 3) which have been included as their values (either only %S or only N/S) fall within the desired ranges when corrected for weighing errors on the analytical balance.The only fish, a garpike, has a C/S ratio of 246, outside the stipulated range of 125-225 for fish.However, this range was established using a dataset that is heavily biased towards cod (>90% of the dataset: 89 out of 98 modern samples and 39 out of 42 historic samples) (Nehlich and Richards 2009).The bones of garpike, unlike cod, have a bluish-green colour, recently identified as biliverdin, associated with higher proportions of hydroxyproline (Jüttner et al. 2013), which could account for the relatively low sulphur concentration and thus elevated C/S ratio.Given the distinctive nature of garpike bone, it is therefore conceivable that the quality criteria for fish suggested by Nehlich and Richards (2009) may not be applicable.The sample has accordingly been included.

Human samples
Observed δ 34 S values for the human samples are presented in Table 3 and Figure

The megalithic tomb in context
The fact that the Resmo passage grave was used during three separate phases, makes it  Only those human individuals from Rössberga, Frälsegården and Alvastra who have been directly radiocarbon dated, and coincide in date with Resmo Phase 1, have been included in the comparison (Figure 4).From Rössberga, the δ 34 S values of bone collagen from eleven individuals range from +9.8 to +12.4‰ (+10.9‰± 0.8‰), indicating an exclusively local origin, based on the limited variation and compatibility with δ 34 S of contemporaneous fauna from the tomb (Linderholm 2008).The fact that the Rössberga data derive solely from bone, could potentially result in less variation than the inclusion of both bone and dentine values.
This, however, does not seem to be the case, as δ 34 S data from the other Falbygden megalithic tomb, Frälsegården, includes both dentine and bone values from eleven individuals (17 samples), with only a slightly larger range, from +9.1 to +12.3‰ (+10.6 ± 0.9‰).Bone and dentine collagen δ 34 S values from eight individuals (14 samples) from the Alvastra dolmen, by contrast, are much more dispersed, ranging from +4.4 to +10.6‰ (+7.6 ± 1.9‰), extending outside the faunal range, most likely including people from a much wider area than the vicinities of Alvastra (Fornander 2011).Although the geology at Alvastra is different from both Rössberga and Resmo, a comparison is nevertheless considered justified (for a detailed argument, see Fornander (Fornander 2011, 134f).
Somewhat surprisingly, the standard deviations of δ 13 C, δ 15 N and δ 34 S are largely similar for Rössberga and Resmo Phase 1, and there is much overlap between their δ 34 S values.It is also noteworthy that although Rössberga has a geology resembling that of Öland, and any sea spray effect can be ruled out, the δ 34 S mean for Rössberga is higher than for Resmo -the mere presence of δ 34 S values >10‰ in Resmo accordingly does not inherently imply a sea spray effect.The Alvastra dolmen stands out with its much higher standard deviations and much less overlap in δ 34 S values with the other two tombs.Thus, although Resmo is the site geographically most distant from the megalithic core area Falbygden, it was apparently not so different with regard to mobility.
In Figure 4, it is also evident that Phases 2 and 3 in Resmo are very different from the earlier phase, both in terms of variation and mean values.Especially Phase 3 seems to match the Alvastra dataset, which suggests that the level of mobility could be of a comparable magnitude.

Modelling the terrestrial δ 34 S
There is a statistically significant positive correlation between δ 13 C and δ 34 S for humans (Spearman's r= 0.55, p<0.001), which is also obvious from Figure 3.The implication is that the marine component of the diet elevates the sulphur isotope value, thus obscuring mobility patterns.In order to discern mobility from such data, it is therefore necessary to isolate the δ 34 S of the terrestrial dietary protein.Eriksson et al. (2013), an estimate of the original terrestrial δ 34 S value can be calculated, which can subsequently be compared to the local terrestrial range.
Previous δ 13 C and δ 15 N analyses (Eriksson et al. 2008) have demonstrated that the Neolithic and Bronze Age humans on Öland relied essentially on only two main protein sourcesterrestrial herbivores and marine mammals.This was based on extensive faunal data showing a clear isotopic separation between fish and marine mammals, where the strong correlation between human δ 13 C and δ 15 N ruled out any substantial contribution of fish protein to the human diet (whereas dogs were suggested as the main consumers of fish, providing a better fit with the isotopic data cf.Eriksson 2004).
Because there are only two main protein sources, and because of the demonstrated association between δ 34 S and δ 13 C, a linear mixing model can be employed (making the use of more complex models superfluous).Accordingly, the estimated terrestrial δ 34 S value, R t , is calculated as: where R obs is the observed δ 34 S value, and R mar is the marine δ 34 S end-point value, +15‰.M is the marine factor, which is calculated as where C mar is the percentage of marine dietary protein, which is calculated from δ 13 C, using -22‰ and -13‰ as terrestrial and marine end-points, respectively.
The calculation of the marine δ 34 S mean (R mar , +15‰) is based on marine mammals only, that is, harbour porpoise (n=3), ringed seal (n=2) and harp seal (n=1).The δ 13 C and δ 15 N evidence shows that the marine protein consumption derived mainly from marine mammals, and apparently not from fish or birds to any large extent.Moreover, the marine bird (+8.4‰) was not determined to species, and therefore whether it was migratory could not be established.This does not exclude the possibility that its flesh or eggs constituted a minor part of the diet, and its value is accordingly informative.Much the same could be said about the garfish (+17.3‰), with the addition that its C/S ratio indicates that it should be treated with caution.Including the bird and the fish would not substantially alter the mean marine δ 34 S value (from +14.8 to +15.2‰), and it was deemed reasonable to exclude them from the calculation.
The local terrestrial range, reflecting the bioavailable δ 34 S composition of the island, is calculated as the δ 34 S mean ± 2 standard deviations of wild local terrestrial fauna.The resulting range, +6.3‰ to +13.7‰, is thus based on δ 34 S data from moose (n=3), roe deer (n=1), pine marten (n=1), mountain hare (n=5) and wild/feral pigs (n=3) (cf.Table 2).All these species are likely to have fed both at the coast and the interior, and their home ranges to have been naturally delimited by, and in most cases covered, the whole island, which is more or less geologically homogeneous.The mountain hare displays the widest range of δ 34 S values, from 7.8 to 13.6‰, probably reflecting more limited home ranges of individual hares.In view of the limited size of the island, it is likely that the terrestrial plants, and by implication also the local fauna, are to some extent influenced by the sea spray effect -the hare displaying the highest terrestrial isotopic value, +13.6‰, is conceivably an example of this.However, the sea spray effect cannot be expected to be as substantial on Öland as on islands of comparable size in the ocean or the Mediterranean, first because of its proximity to the mainland, but above all because the size, morphology, salinity and history of the Baltic Sea differ radically from the oceans.The established local terrestrial range is consistent with these facts, also supported by δ 34 S analysis of modern cod, demonstrating that Baltic cods have lower δ 34 S than Atlantic ones (Nehlich et al. 2013).The impact of the sea spray effect is also considerably lower because of the lower marine δ 34 S range.The application of the model allows the identification of genuine mobility patterns at the individual level.As for any model, there are of course intrinsic uncertainties and errors, and these increase with the percentage of marine protein input.At a certain point there is accordingly so much imprecision that the estimate is no longer informative.Our assessment is that this point is reached at around 60% marine dietary input.Consequently it is not possible to make any estimates for the Köpingsvik individuals.Their δ 34 S values correspond with previously published data from the Pitted Ware site Korsnäs in eastern central Sweden (Fornander et al. 2008), indicating that high marine protein consumers have uniform values throughout the Baltic Proper.Residential changes will accordingly not be discernible.

Human mobility in Resmo
Estimated terrestrial δ 34 S values for human individuals with marine dietary input below 60% were calculated for 28 individuals from Resmo and three from Torsborg, including intraindividual data for 21 Resmo individuals.These are plotted separately for each phase in   In sum, there is an increase in the proportion of individuals classified as non-locals from the Neolithic (Phases 1 and 2), to the Bronze Age (Phase 3) -from 50% and 36% during the Neolithic, to 67% during the Bronze Age.This rise is clearly discernible but by no means remarkable.However, taking into account the overall range of variation for each phase, a clear pattern emerges (Figure 6).The mean estimated terrestrial δ 34 S value for Phase 1 is +6.7‰, and the standard deviation only 1.4‰ (n=15), while both the mean and the sd increase during Phase 2, to +7.7 ± 2.1‰ (n=20, incl.Torsborg).During Phase 3, the mean is below the local terrestrial range, +6.2‰, and the sd as large as 3.3‰ (n=27, incl.Torsborg).
As expected, the subadult values (dentine and subadult bone) vary to a higher degree (both range and sd) than adult values (adult bone), since the former represent shorter tissue formation times, while the longer time of formation in adult bone tends to level out shortterm variation such as seasonal differences.It is truly difficult to find a better explanation for the increasing ranges at the population level than increasing mobility.

Diet and mobility
Dietary life history data was available for the majority of the Resmo individuals.Based on individual ranges of δ 13 C and δ 15 N, their dietary variation was classified as limited, moderate and pronounced, respectively (Eriksson et al. 2008;Eriksson and Lidén 2013).Evidently, four individuals experienced changes in residence not accompanied by any major shifts in diet: subjects 2 and 10 in Phase 1, and subjects 13 and 23 in Phase 3. Five individuals have changed residence while also markedly changing diet (moderate or pronounced dietary variation): subjects 12, 16, 19 and 21 in Phase 2, and subject 18 in Phase 3. Of the twelve individuals for whom no change of residence was discernible, only one, subject 8 in Phase 1, had any notable dietary variation, while the remaining eleven individuals had limited dietary variation: subjects 1 and 6 in Phase 1, subjects 3, 15 and 20 in Phase 2, and subjects 17, 24, 26, 27, 28 and 30 in Phase 3. Consequently, it seems that major shifts in diet could be explained primarily by change of residence.

Archaeological implications
Food and cuisine are strongly associated with cultural identity, and, as such, typically resistant to change.A change of diet therefore suggests that a cultural transformation has taken place.In the case of the Resmo megalith, where no artefacts can be associated with individuals, the only individual cultural indicators besides the megalithic tomb itself are the diet and date.The extended use of the passage grave makes it clear that the cultural affiliation with the TRB cannot be valid for all the interred individuals.Thus, after the initial erection and use in Phase 1, we interpret the changes in diet and mobility as reflecting a major cultural transformation, possibly connected to the appearance of the Battle Axe material culture.Phase 3, by contrast, is characterised by an increasing number of non-local people.These newcomers brought with them intensified agriculture, trade and metal craftsmanship; this is also reflected in the general material culture of the Bronze Age, where we see intensified contacts with continental Europe, e.g., in amber and bronzes (Kristiansen and Larsson 2005).The isotope analysis and the subsequent application of our model provide real insights into human mobility at the individual level -an important addition to studies of mobility based on the presence of exotic artefacts or analyses of population genetics.

Ware
Culture on Öland.The two major sites analysed were the passage grave in Resmo, and the Pitted Ware habitation and burial site in Köpingsvik (for more detailed information about the sites, see e.g.Papmehl-Dufay 2006;Eriksson et al. 2008).All analysed individuals were directly radiocarbon dated, revealing that the megalithic tomb in Resmo was in use during three phases.The first phase, c. 3500-2900 BC, can be attributed to the TRB presence on the island, and the diet during this period is characterised by a mixture of marine and terrestrial protein sources.During the following phase, c. 2900-1900 BC, the dietary components are still the same, but substantial inter-and intra-individual differences in the proportions of marine vs terrestrial protein are evident in the stable isotope data.The third phase, c. 1900-1000 BC, is characterised by a seemingly complete reliance on terrestrial (probably domesticated) resources (see furtherEriksson et al. 2008).At the Köpingsvik site, located less than 50km from Resmo, the majority of the analysed individuals overlapped chronologically with Resmo Phases 1 and 2, c. 3300-2500 BC, displaying a diet dominated by marine mammal protein.At a third site, the Torsborg gallery grave complex, ranging in date from the Middle Neolithic to the Early Bronze Age, c. 2900-1400 BC, the dietary patterns correspond to their chronological equivalents at Resmo during Phases 2 and 3, respectively.
3. The 70 samples derive from 36 individuals, and include intra-individual data for 22 of these individuals.The δ 34 S values for six Middle Neolithic individuals from Köpingsvik range from +13.3 to +15.3‰ (+14.0 ± 0.7‰, n=7), very similar to the range for marine mammals, and consistent with a diet of predominantly marine mammal protein.The only intra-individual data, for Grave Klinta A7, demonstrate no change in δ 34 S from childhood to adult age.

Figure 3 :
Figure 3: Observed δ 13 C, δ 15 N and δ 34 S values for all human samples subjected to δ 34 S analysis.
necessary to consider the cultural context for each phase separately.Phase 1 in Resmo is linked to the erection of the megalithic tomb and its first use, and can on good grounds be associated with the Funnel Beaker Culture (TRB).Comparison with δ 34 S data from three other Swedish megalithic tombs may help put the Resmo data into a wider context.Data are available for the Rössberga and Frälsegården passage graves in Västergötland, both situated in the Falbygden area, and for the Alvastra dolmen in Östergötland -a monument which, together with the Resmo tombs and the Ansarve Hage dolmen on Gotland are isolated phenomena in the Swedish megalithic world, as they are the only ones located outside the core areas of Falbygden and the coasts of Bohuslän, Halland and Scania.The megalithic tombs at Resmo and Alvastra are consequently both unique features that can be expected to deviate from the Rössberga and Frälsegården tombs with regard to mobility patterns.

Figure 4 :
Figure 4: Observed δ 34 S, δ 13 C and δ 15 N values (mean ± 1 sd) for Resmo humans, grouped according to phase.Three megaliths on the Swedish mainland are also shown for comparison.Only radiocarbon dated individuals coinciding with Phase 1 in Resmo were included from the Rössberga passage grave, the Frälsegården passage grave and the Alvastra dolmen.
Domestic species were not included in the calculation of the local terrestrial range, mainly because of the risk that they were imported, or subject to specific cultural practices affecting what they fed on, hence resulting in δ 34 S values not representative of the bioavailable sulphur of the local environment.The Middle Neolithic cattle specimen from Köpingsvik -clearly contemporaneous with the Pitted Ware population, but culturally an 'exotic' -is an example of the former, dogs of the latter.Nevertheless, the majority of the domestic animals -including one pig from Resmo -display rather homogeneous δ 34 S values, within the local terrestrial range, consonant with a local origin.This is not surprising, as it is highly unlikely that people in Neolithic and Bronze Age Scandinavia relied to any large extent on imported livestock.Two specimens from Resmo fall below the local terrestrial range; an ovicaprid tibia of historical date (thus probably imported) and a pig tooth (not dated).

Figure 5 ,
Figure5, where the predicted local terrestrial range is shaded, black circles around symbols mark adult bone samples, black squares around symbols mark subadult bone samples, and the remaining symbols are dentine samples.It is important to remember here that the estimated terrestrial δ 34 S value reflects only the terrestrial portion of the diet, which enables comparison between individuals and chronological phases with varying levels of marine dietary protein input.

Figure 5 :
Figure 5: Estimated δ 34 S values of the terrestrial food components for human samples, plotted against calibrated radiocarbon age (centre value of 2σ calibrated range) during Phase 1, 2 and 3,

Figure 6 :
Figure 6: Estimated terrestrial δ 34 S values for Resmo and Torsborg for each chronological phase.Crosses include all dentine and subadult bone values, while circles include only adult bone values.

Table 1 :
Chronological periods for southern Sweden

Table 4 :
Overview of the interpretation of individual dietary and mobility isotope data for Resmo (subjects 1- Stable sulphur isotope data for the four individuals from Torsborg (one sample each) range from +2.6‰ to +13.6‰.The Middle Neolithic individual, the only one with a carbon isotope

Table 2 :
Eriksson et al. 2008)esmo (RES), Köpingsvik (KOP) and Torsborg (TOR) successfully analysed for δ 34 S, sorted according to species.Precision for δ 13 C and δ 15 N ±0.15‰ or better for all samples (data fromEriksson et al. 2008).Precision for δ 34 S ±0.3‰ or better at Iso-Analytical (ISO) and ±0.2‰ or better at SIL. *= %S or N/S falls within the desired ranges (cf.Nehlich and Richards 2009)when corrected for weighing errors on the analytical balance.**=C/S outside range for fish, but see main text (Section 4.1)

Table 3 :
Eriksson et al. 2008)smo (RES), Köpingsvik (KOP) and Torsborg (TOR) successfully analysed for δ 34 S, sorted according to site and individual.Precision for δ 13 C and δ 15 N ±0.15‰ or better for all samples (data fromEriksson et al. 2008).Precision for δ 34 S ±0.3‰ or better at Iso-Analytical (ISO) and ±0.2‰ or better at SIL. NB: Radiocarbon dates are indicated for each individual, not for samples *= %S or N/S falls within the desired ranges (cf.Nehlich and Richards 2009)when corrected for weighing errors on the analytical balance ¤ =%C corrected by a factor of 1.15 because of an elemental analyser error during one run, seeEriksson et al. 2008, 529, table 3, for details

Table 5 :
Human and faunal samples which fail to meet the quality criteria for sulphur isotope analysis with regard to %S, C/S or N/S(Nehlich and Richards  2009), and are accordingly excluded.None of the samples analysed from the sites of Kalleguta (KAL), Vickleby (VIC) and Algutsrum (ALG), comply with the quality criteria and they are therefore not discussed in the main text