Eagle Hill Masthead

Journal of the North Altantic

    Aim and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Co-published Journals
    Northeastern Naturalist
    Southeastern Naturalist
    Caribbean Naturalist
    Neotropical Naturalist
    Urban Naturalist
    Eastern Paleontologist
    Eastern Biologist

Eagle Hill Institute

About Journal of the North Atlantic


Contrasting Patterns of Resource Exploitation on the Outer Hebrides and Northern Isles of Scotland during the Late Iron Age and Norse Period Revealed through Organic Residues in Pottery
Lucy J.E. Cramp, Helen Whelton, Niall Sharples, Jacqui Mulville, and Richard P. Evershed

Journal of the North Atlantic, Special Volume 9 (2015): 134–151

Full-text pdf (Accessible only to subscribers.To subscribe click here.)


Site by Bennett Web & Design Co.
Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 134 Introduction Recent investigations of archaeological remains from the Northern Isles of Scotland have indicated that the Viking incursions and later Norse settlement on the islands led to novel patterns of economy and subsistence. Although debate continues regarding the nature of the early interaction between the native Picts and these Viking incomers (Barrett et al. 2000, Crawford 1981, Morris 1996, Ritchie 1974), novel settlement patterns and more intensive marine exploitation, leading to eventual trading and exportation of marine resources, are clearly visible (Barrett 1997; Barrett and Richards 2004; Barrett et al. 2000, 2001; Cerón-Carrasco 2005; Graham Campbell and Batey 1998:214–215) and are testimony to Scandinavian influence on the islands. On the Outer Hebrides, on the other hand, it has been argued that settlement patterns succeeding the Pictish period do not demonstrate such profound changes to Iron Age settlement patterns, with some level of continuity postulated as opposed to a swift and violent population replacement (Sharples and Parker Pearson 1999; though see Sharples and Smith 2009). Nonetheless, changes in economic patterns, namely increased and more specialized marine product exploitation during the Norse period are likely (Smith and Mulville 2004:55). Here we provide new evidence for dietary and economic patterns during the Late Iron Age and Norse period by exploiting biomarkers preserved in absorbed and visible residues from a large number of pottery sherds excavated from Late Iron Age and Norse contexts from Bornais, on South Uist (131 vessels), and Late Norse pottery from Jarlshof, on Shetland (27 vessels). This approach offers a very high level of sensitivity for the detection of marine product processing and furthermore allows the exploitation of secondary (e.g., dairy) products to be directly demonstrated. These new findings demonstrate distinct patterns of resource processing in pottery at these two island locations. Investigation of Organic Residues in Archaeological Pottery The investigation of preserved ancient biomolecules can offer new insights into the substances processed or contained in unglazed vessels. This knowledge, in turn, can lead us to answer questions of wider archaeological significance, including resource exploitation strategies, long-distance trade links, cultural practices, and long-term dietary patterns. It is well established that lipid components become absorbed and protected within the ceramic matrix of unglazed pottery, with a negligible contribution of lipid from the burial environment (Heron and Evershed 1991). After the extraction of biomolecules using organic solvents, diagnostic components in these residues can be related to distributions identified in modern reference plant and animal products. However, the specificity of identification is dependent upon the degree of uniqueness of these biomarkers (biological markers) to particular resources, and the persistence of these or their degradation products (Evershed 1993, 2008). Hence, while beeswax, for example, contains characteristic distributions of palmitate wax esters (e.g., Evershed et al. 1997), it is often difficult to distinguish plant waxes of different origins due to the ubiquitous nature of many plant wax components (Kolattakudy et al. 1976, Tulloch 1976). On the other hand, Contrasting Patterns of Resource Exploitation on the Outer Hebrides and Northern Isles of Scotland during the Late Iron Age and Norse Period Revealed through Organic Residues in Pottery Lucy J.E. Cramp1,2,*, Helen Whelton1, Niall Sharples3, Jacqui Mulville2, and Richard P. Evershed1 Abstract - This paper presents the findings from an investigation of organic residues extracted from pottery sherds from Late Iron Age and Norse phases from Bornais, South Uist, and the Late Norse period from Jarlshof on Shetland. These data confirm intensive and/or specialized processing of marine products in pottery on Shetland, either for consumption or other uses, such as rendering of oil from fish livers. In contrast, at Bornais, little increase in the intensity of marine product exploitation can be identified between the residues from the Later Iron Age and Norse phases; however, an emphasis on dairy products is identifiable throughout all phases and pottery types. While the findings from these two sites clearly cannot be extrapolated as entirely representative of the wider respective regions, what emerges is further evidence for diverse economic or cultural patterns at different locations within Scandinavian Scotland. Special Volume 9:134–151 2010 Hebrideian Archaeology Forum Journal of the North Atlantic 1Organic Geochemistry Unit, School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, UK. 2Current address - Department of Archaeology and Anthropology, University of Bristol, 43 Woodland Road, Bristol BS8 1UU, UK. 3SHARE, Cardiff University, Colum Drive, Cardiff CF10 3EU, UK. *Corresponding author - lucy.cramp@bristol.ac.uk. 2015 Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 135 although modern dairy fats can be distinguished from adipose fats with relative ease due to high abundances of short-chain fatty acyl components in milk fats, the preferential hydrolysis and loss of shorter-chain components means such distributions are rarely well preserved in an archaeological fat residue. As such, it is necessary to investigate the stable carbon isotopic composition of individual saturated fatty acid components in order to separate milk from adipose fats, and these from non-ruminant fats (Dudd and Evershed 1998, Dudd et al. 1999, Evershed et al. 2002). Organic residues of aquatic origin are a strong likelihood at later prehistoric coastal and island sites and are a particular challenge to identify in archaeological lipid extracts. This difficulty is due to high abundances of long-chain polyunsaturated fatty acids (PUFAs), such as eicosatetraenoic acid (C20:4), eicosapentaenoic acid (EPA; C20:5) and docosahexaenoic acid (DHA; C22:6) in marine and freshwater lipids (Fig. 1A). These highly unsaturated components are susceptible to rapid oxidation and hence, loss, from the original fatty acid distribution, resulting in a fatty acid distribution dominated by the more stable saturated components, thus closely resembling degraded terrestrial fats (Fig. 1B). It is therefore necessary to exploit more stable products that derive from the presence of double bonds in the precursor fatty acids and, importantly, preserve the carbon chain length of the original fatty acids. Oxidation products, including mono- and dihydroxy fatty acids, arise through the hydroxylation of double bonds in monounsaturated fatty acids, and the carbon chain length of these products therefore reflects that of the precursor fatty acid, while the position of the hydroxyl groups preserves the position of the original double bonds. Chemical bonding or adsorption of these components into an insoluble polymeric matrix means it is often necessary to perform an alkaline extraction in order to release the “bound” lipids from the pottery fabric. These derivatives have been successfully extracted from archaeological pottery that was used to process plant and marine products that would have originally contained significant quantities of monounsaturated fatty acids (Copley et al. 2005, Hansel and Evershed 2009, Hansel et al. 2011, Regert et al. 1998), and the detection of longer-chain dihydroxy acids (>C20) is diagnostic of an aquatic source. Further evidence for a marine contribution to an archaeological residue can be sought through the Figure 1. Partial GC/MS total ion and mass chromatograms showing (A) the fatty acid composition of fresh limpet and structure of C20:4 (n-6) fatty acid and (B) the fatty acid composition of limpet fat degraded in the laboratory at 270 °C for 17 h. (C) m/z = 105 selected ion monitoring (SIM) mass chromatogram, which is the fragment ion common to all APAAs, showing distributions of C18 (M+.290), C20 (M+.318) and C22 (M+.346) ω-(o-alkylphenyl)alkanoic acids present in (B) and the structure of a C20 APAA (n-0). Fatty acids were analyzed as their methyl esters. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 136 identification of ω-(o-alkylphenyl)alkanoic acids (APAAs) that form from the thermal degradation of polyunsaturated fatty acids (Matikainen et al. 2003). We have demonstrated via heating experiments of pure fatty acid standards and mixtures of fatty acids that the carbon chain length of these products reflects that of the precursor fatty acids (Cramp and Evershed 2014, Evershed et al. 2008), and therefore the detection of C20 and C22 APAAs in archaeological residues is evidence that products containing longchain polyunsaturated fatty acids (i.e., marine fats and oils) were processed in the pottery (Copley et al. 2004, Evershed et al. 2008, Hansel et al. 2004). APAAs are only likely to be present in extremely low concentrations, but by operating the mass spectrometer in selected ion monitoring (SIM) mode, it is possible to detect these diagnostic compounds at very high sensitivity (Fig. 1C). In addition, a third class of biomarker, isoprenoid fatty acids, is also a diagnostic of marine fats and oils. This class of compounds is not a product of degraded unsaturated fatty acids but rather derives from phytol in marine algae, accumulating in organisms further up the food chain. The presence of one or more of the isoprenoid fatty acids 4,8,12-trimethyltridecanoic acid (4,8,12-TMTD), 2,6,10,14-tetramethylpentadecanoic acid (pristanic acid), and 3,7,11,15-tetramethylhexadecanoic acid (phytanic acid) may therefore be observed in residues of a marine origin. However, since at least some of these isoprenoid fatty acids (e.g., phytanic acid) are known to also exist in terrestrial fats such as milk (Vetter and Schröder 2011), these are not robust biomarkers in isolation. At the Late Iron Age and Viking/Norse island settlements investigated here, likely resources include domesticated and wild ruminants (including cattle, sheep, and red deer), some pig, marine shellfish, fish, marine mammals, cultivated cereals, and other wild and domesticated plants. Through the investigation of distributions of biomolecules, including the use of SIM to detect those present only in low concentrations, in combination with the determination of stable carbon isotope compositions of individual fatty acids, it is possible to determine the origins of the major sources of lipids recovered from this pottery in addition to some of the more minor commodities that may have been more sporadically processed. The aim of the research presented here was to reconstruct key aspects of these island economies and, especially, to investigate the changing importance of marine products (e.g., fish, shellfish, marine mammals) from the Late Iron Age to the Norse period, through the investigation of biomarkers preserved in the cooking pottery used by the inhabitants of these settlements. Pottery Investigated The settlement at Bornais is located on the machair plain of the western coast of South Uist, which forms part of the island chain of the Outer Hebrides off the western coast of Scotland (Fig. 2). It is a multiphase site, first settled during the Middle Iron Age (AD 200–400) and inhabited throughout the Late Iron Age (5th–8th century AD), Viking (9th–10th century), and Norse period (10th–14th century), showing a decline by the 15th century AD. The site comprises three large mounds (1, 2, and 3) and two subsidiary mounds (2A and 2B). Exploratory excavations were undertaken on mounds 1 and 3, while mound 2 and mound 2A have seen substantial area excavations (Sharples 1999, 2000, 2005; Sharples and Smith 2009). Although all mounds produced evidence for Norse activity, a Late Iron Age wheelhouse was excavated on mound 1. Mound 2 appears to have been the center of the settlement and revealed a sequence of three substantial buildings, including a large bowshaped hall dating to the 11th century. While mound 2A also contained a sequence of buildings, the focus of this excavation was upon the associated ancillary structures and middens. The excavations at Bornais produced rich cultural and ecofactual material, including worked bone objects, pottery, carbonized plant remains, and faunal assemblages. Objects including steatite bowls, glass beads, metalwork, and walrus ivory are indicators of trade links with Shetland, Scandinavia, England, and Ireland (Sharples 2000). However, although ceramic forms underwent a significant change with the Norse period, it has been argued that the continued use of ceramics per se and the similarity between Late Iron Age and Viking settlement patterns and architectural features suggest a significant continuity of the indigenous Pictish population during the Viking age (Sharples and Parker Pearson 1999). From the large pottery assemblages excavated, 131 sherds and any associated carbonized residues from Mounds 1, 2, 2a, and 3 were investigated. These sherds date from the Late Iron Age through to the Late Norse period and, as such, were grouped according to phase, comprising Late Iron Age I (5th–6th century AD), Late Iron Age II (Pictish; 7th–8th century AD), Early Norse (10th–early 11th century AD), Mid-Norse (Late 11th– 12th century AD), and Late Norse (13th–14th century AD) vessels. Sherds were selected from the major vessel forms, including jars and bowls. Alongside these probable “cooking pots”, 13 distinctive Middle and Late Norse “platters” were included for comparison. While also colonized by the Vikings, the Northern Isles of Scotland, including Shetland and Orkney, and the northeast mainland of Scotland, comprised a region that can be distinguished in Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 137 chronology, environment, and culture from the Outer Hebrides, with persuasive evidence for the Viking introduction of a strongly marine-based subsistence strategy and economy (Barrett et al. 1999, 2001). The Pictish period on the Northern Isles is succeeded by the Viking Age at ca. 800 AD, with the Norse period continuing from the 11th– 14th century AD. Jarlshof, located near Sumburgh Head on the southern tip of Shetland (Fig. 2) has revealed extensive Viking and Norse remains which succeed substantial prehistoric activity. The earliest evidence dates to the Neolithic, and occupation continued through the Bronze and Iron Ages to the numerous rectilinear structures that are associated with Scandinavian settlers. The Jarlshof site continued to be inhabited during the Medieval and post- Medieval era until the 17th century when the large “Laird’s House” was abandoned. Although at Jarlshof the Viking and Norse remains are extensive, it is unlikely this settlement was ever a large village. It is more probable that the excavated buildings were occupied during successive phases over several centuries, each of which saw new modifications and additions to the settlement plan. Rich cultural remains were excavated, and yet because systematic wet sieving was not undertaken, the environmental remains are unfortunately biased, and limited, due to the nature of their recovery. With a preference for steatite, rather than ceramic vessels during the Viking period at Jarlshof, pottery only re-emerged relatively late in the sequence of Norse activity (Hamilton 1956:188, Figure 2. Map of the British Isles showing locations of sites d iscussed in the text. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 138 Ritchie 1977). The pottery analyzed here therefore mainly derived from phases VII and V (approximately 12th–13th century AD; Hamilton 1956), and is probably contemporaneous with the pottery from the Mid- and Late Norse period at Bornais, which spans the 11th to 14th century. This pottery is relatively crude, grass-tempered hand-made ware, including Hamilton’s “Class II”, described as a “large open bowl” (Hamilton 1956:188). The sherds selected are described in Table 1, which includes a summary of the phases and forms included in the study. Where possible, sherds were selected from the upper body of the vessel, since this part of the vessel is likely to contain the highest concentrations of lipid (Charters et al. 1993, 1997). Thick visible carbonized encrustations were present on the inner and outer surfaces of a large number of the sherds. Portions of these residues were also removed and analyzed alongside the absorbed residues. The analytical and instrumental protocols are described in Appendix 1. The istotope and marine biomarker data for all of the sherds found to contain significant quantities of lipid are presented in Appendix 2. Findings Late Iron Age to Late Norse pottery from Bornais (AD 400–1400) Organic residues were well preserved in the absorbed and visible residues from Bornais, with appreciable concentrations of lipid extracted from 60% of sherds. No lipids were extractable from the Mid- and Late Norse platters included in the study, suggesting that these vessel forms were used for lipid-poor commodities, such as bread. Otherwise, the majority of residues from Bornais were characterized as degraded animal fat due to high relative concentrations of stearic acid (C18:0 fatty acid) and characteristic distributions of tri-, di-, and monoacylglycerols (Fig. 3). A wide range of triacylglycerol carbon numbers, with distributions closely resembling partially degraded dairy fats (Christie 1978, 1981; Dudd et al. 1998; Evershed et al. 2002), was observed in many sherds from the Late Iron Age through to the Late Norse period. The determination of stable carbon isotope values of individual fatty acids from 78 residues confirms that, in fact, 35% (n = 11/31) of Late Iron Age I and II residues and over 50% (n = 24/47) of Early to Late Norse residues from Bornais lie within the range expected for dairy fats (Figs. 4, 5; Copley et al. 2003 ). Dairy fats could not be related to any specific vessels, being present in all identifiable types (excepting the platters), including open and convex bowls and jars. The analysis of faunal remains from Bornais indicates that ruminants, in particular, cattle, sheep, and deer, were a significant contributor to the diet at Bornais, and the high number of neonatal domesticates observed, particularly among cattle assemblages, is indicative of a dairying economy in both the Late Iron Age and Norse periods (Mulville 1999). The faunal bone age indicators produced conflicting interpretations regarding whether the highest frequency of neonates lay in the Late Iron Age or Norse periods (Mulville 1999), with dental evidence suggesting greater emphasis in the Late Iron Age compared with the Norse period, and bone-fusion evidence suggesting the converse. The residues analyzed here are certainly supportive Table 1. Identities of the sherds investigated from each major phase of the settlement at Bornais and Jarlshof, noting any characteristic forms. No. analyzed Visible No. δ13C Phase Sherds residues determinations Forms identified Bornais Late Iron Age (5th–8th century AD) 4 2 0 Jars (n = 3; globular and shouldered); bucket (n = 1) Late Iron Age I (5th–6th century AD) 20 3 6 Shouldered jar (n = 3) Late Iron Age II (7th–8th century AD) 25 20 25 Total 49 25 31 Early Norse (10th–11th century AD) 25 10 19 Open -mouth bowl (n = 1), platter (n = 1) Middle Norse (Late 11th–12th century AD) 25 9 13 Platters (n = 7), convex bowls (n = 3), open-mouth bowls (n = 3) Mid-Late Norse (Late 11th–14th century AD) 4 - 4 Late Norse (13th–14th century AD) 24 6 9 Platters (n = 3), convex bowls (n = 2) Norse (10th–14th century AD) 4 1 2 Platters (n = 3) Total 82 26 47 Jarlshof Late Norse, House 5 (12 th–13th century AD) 10 4 8 Square mouthed bowl (n = 1) Late Norse, House 7 (12 th–13th century AD) 12 - 11 Round bowls (n = 12) Late Norse (12th–13th century AD) 5 1 3 Total 27 5 22 Total 158 56 100 Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 139 30% (n = 20) of Early Norse, 13% (n = 15) of Mid- Norse, 50% (n = 4) of Mid–Late Norse and 28% (n = 18) of Late Norse vessels investigated using SIM. Long-chain dihydroxy fatty acids were identified in ≈40% of “bound” lipids subsequently extracted from these sherds (Figs. 4, 6). These findings confirm that, in addition to ruminant products, marine resource exploitation can be identified in both the Late Iron Age II and subsequently throughout the Norse period. Although a broad increase in the frequency of marine biomarkers between the Late Iron Age and Norse phases can be observed, there is actually only a slight increase in evidence for marine product exploitation between the Late Iron Age II and the Early Norse period, and there is not an increasing trend in the frequency of marine biomarkers throughout the Norse period. Thus, there is no clear association of marine product intensification with the beginning of the Viking age. of intensive exploitation of dairy products during both the Late Iron Age and the ensuing Norse period, and the higher frequency of residues of dairy origin during the Norse period lends greater support to the bone-fusion rather than the dental evidence. Although the compound-specific stable carbon isotope values indicate that predominantly terrestrial resources were prepared in the pottery vessels from all phases investigated at Bornais, substantial quantities of marine fat would be required to significantly shift these values, such that lower quantities of aquatic products may be undetectable. Therefore, solvent- and alkaline extracted residues were analyzed using GC/MS operating in SIM mode in order to detect other marine biomarkers that might be present only in very low concentrations, including ω-(oalkylphenyl) alkanoic acids (APAAs) and dihydroxy acids. Although no long-chain APAAs were detected in the Late Iron Age I pot residues (n = 8), these were identified in 20% (n = 25) of the Late Iron Age II, Figure 3. Partial high-temperature gas chromatograms from trimethylsilylated lipid extracts from the sherd fabric and its carbonized deposit from Bornais. X:Y FA–free fatty acid, where X is the number of carbon atoms and Y the degree of unsaturation. X MAGs–monoacylglycerols, X DAGs–diacylglycerols, X TAGs–triacylglycerols, with X number of acyl carbon atoms. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 140 Figure 4. Scatter plots of the δ13C values of the major fatty acids (C16:0 and C18:0) in the archaeological lipid extracts from Bornais, Late Iron Age I (a), Bornais, Late Iron Age II (b), Bornais, early Norse (c), Bornais, Mid-Norse/Mid–Late Norse (d), Bornais, Late Norse (e), and Jarlshof, Late Norse (f). Note the offset of the lipid extracts from the reference ellipses, resulting from a marine signal causing enrichment in values. These solid ellipses are derived from modern UK terrestrial and marine fauna and are corrected for the effect of the release of δ13C-depleted post-industrial carbon CO2 (+1.2 ‰; Friedli et al. 1986). Since atmospheric carbon may not yet be fully mixed in deeper oceans, it is possible some deep-ocean feeders have values closer to the uncorrected values. indicates where biomarkers indicative of a marine origin were also detected in the residue. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 141 Later Norse pottery from Jarlshof (AD 1100–1300) Organic residues were particularly well-preserved in the absorbed and visible residues from Jarlshof, with appreciable concentrations of lipid extracted from >70% of sherds and 60% of carbonized deposits. The TLEs were usually dominated by the C16:0 and C18:0 fatty acids (Fig. 6); however, in contrast to lipid extracts from Bornais, the C16:0 fatty acid was the major component in the majority of residues. Traces of long-chain (≥C20) monounsaturated fatty acids were also widely present in low concentrations. Based upon the fatty acid composition alone, a high proportion of these residues do not resemble terrestrial animal fats. Fifteen of twenty-two sherd residues analyzed using SIM (68%) contained long-chain APAAs and long-chain dihydroxy fatty acids (Figs. 4, 5, 6). Isoprenoid fatty acids were present in some residues containing other marine-derived biomarkers, albeit in low concentrations; thus far, all three of the diagnostic isoprenoid acids have been identified in only one residue from Jarlshof (Figs. 7,8). The stable carbon isotope values of individual fatty acids were also determined from these 22 residues, with approximately half yielding δ13C values that are more enriched than anticipated for purely terrestrialderived fats (Fig. 4), providing further confirmation of a marine contribution. However, while some residues exhibit values consistent with a pure marine fat origin, approximately half plot along a mixing line, indicating some contribution of terrestrial fats to a mixture. Plotting the Δ13C (δ13C18:0–δ13C16:0) values, which removes effects of climate and environment on absolute values, emphasizes the physiological and metabolic differences between the animals producing the fats (Fig. 5). When plotted in this way, twelve of these residues could be characterized as reflecting a contribution of ruminant fat, with two residues assigned to a dairy fat source, and the remainder to carcass fats. It cannot be discounted that the latter result from the mixing of dairy and marine fats, resulting in Δ13C values that lie within the intermediate ruminant carcass fat range. Although the ecofactual evidence from Jarlshof is sparse and little discussed in the Figure 5. Plot of the difference between the δ13C values of C18:0 and C16:0 (Δ13C) which serves to classify the archaeological lipid extracts from Jarlshof (a) and Bornais (b). The ranges on the y axis are derived from modern reference fats. Stars indicate where aquatic biomarkers were also detected. The largest squares show the mean Δ13C and δ13C16:0 for a dataset of modern marine fauna from the North Atlantic (n = 64; Cramp and Evershed 2014) and Neolithic ruminant carcass and dairy fats. Due to the island location, which is known to result in more enriched terrestrial stable carbon isotope values compared with non-island locations (Farquhar et al. 1989, Heaton 1999), the isotope values from routinely used UK datasets of modern terrestrial fauna (Copley et al. 2003) are likely to have absolute δ13C16:0 values that are too depleted for direct comparison here, although the Δ13C proxy, which reflects differences in metabolism, is known to be largely unaffected by environmental variability (Dunne et al. 2012). Therefore, more appropriate end points for ruminant carcass fats and ruminant dairy fats are derived from a large dataset of values from Neolithic pottery residues shown to contain no contribution of aquatic fats, with Δ13C values of <-3.4‰ (dairy fats, n = 139) and Δ13C values of -3.4 to 0‰ (carcass fats, n = 78). Error bars show one standard deviation. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 142 dominantly terrestrial-derived origin. Since there is a relatively low abundance of stearic (C18:0) acid in marine fats and oils, a low contribution of ruminant adipose or dairy fat (i.e., <25%) would significantly shift the δ13C18:0 values of a marine fat towards a terrestrial signature, as modelled in Fig. 5. We would suggest, therefore, that mixing with terrestrial fats in the majority of vessels was relatively minor to retain generally elevated δ13C and Δ13C values. The high prevalence of aquatic biomarkers and enrichment of stable carbon isotope values leads us to conclude that in contrast to both Late Iron Age and Norse pottery vessels from Bornais, marine excavation report, it has been recognized that the bones of very young animals were relatively frequent, and reassessment of these remains has suggested a dairying strategy—a pattern that has been identified at other Late Norse sites on Shetland such as Sandwick (Bigelow 1985). These data presented here do not entirely contradict this interpretation; however, compared with the residues from Bornais, there is a less noticeable emphasis on dairy products in these particular vessels from Jarlshof, perhaps arising from considerably greater processing of marine products. Only three residues displayed δ13C values and biomarker evidence that was consistent with a pre- Figure 6. Gas chromatograms showing typical fatty acid distributions from Jarlshof and (inset) the identification of three isoprenoid fatty acids in this residue. Fatty acids were analyz ed as their methyl esters. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 143 Discussion Faunal evidence from the Outer Hebrides has suggested that an increase in the intensity of, and investment in, marine product exploitation is associated with the advent of the Viking period (Ingrem 2005, Smith and Mulville 2004). However, at Bornais, where pottery sherds investigated span the Late Iron Age through to the end of the Norse period, the transition to an increasingly marine-based economy is not strongly supported by the findings. Where identified, biomarkers of aquatic origin most frequently occurred mixed with terrestrial products. This finding is significant, implying that at Bornais cooking pots were relatively versatile or, at least, that these different ingredients were prepared together, with almost no evidence for specialized marine-product vessels. Although a broad increasing trend can be identified when comparing the Late Iron Age and Norse periods, the frequency of marine biomarkers in the Middle Norse phase is actually lower than that observed during the Late Iron Age II, and therefore no clear shift towards an increasingly marine-based economy is observed. The large number of organic residues and range of pottery types investigated from Bornais, comprising the further analysis of 93 residues from >150 vessels, suggests that this is representative of a wider phenomenon, whereby the exploitation of marine products in pottery vessels was not intensive, and increased little during the Norse period. resources were widely and intensively processed in pottery at Jarlshof, both exclusively in some vessels and occasionally along with the processing of terrestrial products in other pots. Figure 7. GC mass chromatograms showing the relative intensity of ions diagnostic of C18-C22 dihydroxy fatty acids. For the positive identification of C18, C20, and C22 dihydroxy acids, identical chromatographic peaks must be present for all of the following ions: C18: m/z 215, 317, and M[-15]+ 517; C20: m/z 215, 345, and M[-15]+ 545; C22: m/z 243, 345, and M[-15]+ 573. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 144 Figure 8. GC mass chromatograms showing the relative intensity of ions diagnostic of (C18- C22 ω-(o-alkylphenyl)alkanoic acids. For the identification of ω-(o-alkylphenyl)alkanoic acids, identical chromatographic peaks must be present for the following ions: C18: m/z 105 and M+.290; C20: m/z 105 and M+.318; C22: m/z 105 and M+.346. Although this might appear at odds with the faunal data, this pattern may perhaps be explained in the context of the cultural and economic context in this region. Firstly, analysis of faunal remains from Bornais have suggested that novel subsistence strategies introduced in the Norse period included intensive herring exploitation, a species uncommonly recovered from Norse sites on the Northern Isles (Ingrem 2000, 2005:157). This finding leads us to speculate whether herring specialization may have resulted in the low visibility of marine products in organic pottery residues from Norse-period pottery from Bornais. There is some evidence from Bornais that herring were preserved (Ingrem 2005:176), for example by salting, drying, or smoking. If hung to dry or smoke, this may have left little trace in pottery residues, while soaking in brine is traditionally performed in wooden barrels. The pottery residues from the Norse phase at Bornais would not reveal a shift in fishing practices if aquatic products were being processed using such methods, and it is possible, therefore, that an increase in marine resource processing is actually under-represented. Secondly, a higher degree of population continuity in this region, as argued by Sharples and Parker Pearson (1999), would also explain a less emphatic shift in dietary and economic practices. Evidence from northern Scotland, on the other hand, suggests profound palaeoeconomic changes occurring at the beginning of the Viking Age, moving from relatively low importance of marine products in the late Iron Age (Barrett and Richards 2004) towards deep-water fishing and greater exploitation of large gadids (Barrett 1997; Barrett et al. 1999, 2001; Bigelow 1985; Hamilton 1956; Nicholson 1998; Ritchie 1977), with representation of mammal bones decreasing respectively (Barrett et al. 1999). Deposits likely relating to more intensive fishproduction sites, possibly for trade and export, alongside increasingly Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 145 enriched stable isotope values from Viking and Early Medieval bone collagen from Orkney, suggest a second phase of intensification in fish exploitation from the 11th–14th centuries AD (Barrett and Richards 2004; Barrett et al. 1999, 2000). Also of note is that the preference of species on the Northern Isles shifted towards larger, deep-sea fish such as cod, saithe, and ling (e.g., Barrett et al. 2001, Bigelow 1985, Nicholson 1998, Richie 1976), which were more likely butchered and cooked in pots than smaller fish such as herring, thus increasing their visibility in pottery residues. Thus, it is perhaps unsurprising that the lipid residues from Jarlshof on Shetland can readily be distinguished from contemporaneous vessels residues from Bornais Mid- and Later Norse vessels. The lipid extracts exhibit more enriched stable carbon isotope signatures alongside a very high prevalence of well-preserved aquatic biomarkers. Dating to the 12th–13th century AD, this pottery dates closely to the “fish event horizon” on Orkney, whereby reintensification of marine product exploitation has been inferred at ca. 11th–12th centuries AD through faunal remains and stable carbon isotope values, likely arising from trade and export of dried fish (Barrett 1997, Barrett and Richards 2004, Barrett et al. 2000). There is convincing evidence for the involvement of the Northern Isles of Scotland in the North Sea fish trade at this time, although it is not directly mentioned in historical records (Barrett et al. 2000). Although the drying and/or export of large amounts of fish might not leave direct traces in these pottery vessels, it is still to be expected that, in the context of significant intensification associated with commercial fishing and the processing of large fish on an industrial scale, marine products would likely be readily available, particularly to those involved in the trade. Additionally, preparation of marine products for non-dietary purposes, for example, for lighting fuel, should be considered here. For example, boiling fish livers for lamp oil has been hypothesized from fragments of Nose-period vessels identified at Underhoull on Shetland (Small 1966:242) and ash deposits from St. Boniface on Orkney (Cerón- Carrasco 1994). It is possible that certain vessels from Jarlshof, specifically those with exclusively marine-derived isotope signatures, were used in this role, whether on a domestic or industrial scale. On the other hand, this would imply that at Bornais rendering marine oil through boiling in pottery vessels was not commonplace until post-Medieval times when rendering fish livers in cooking pots is known to have been practiced on the Western Isles (e.g., McGregor 1880). Non-ceramic vessels that might have been used as an alternative are not known at Bornais, and therefore, while wood is a possibility, it is also likely that another substance such as ruminant tallow was collected or rendered for lighting purposes on the Outer Hebrides, as identified in Medieval lamps from Southern England (Mottram et al. 1999). Conclusions The findings presented here lead us to conclude that there are explicit differences between the residues investigated from the Outer Hebrides and the Northern Isles of Scotland, the most obvious of which is the high prevalence of marine-derived biomarkers and marine isotopic signatures in residues from Late Norse pottery from Jarlshof on Shetland. This finding contrasts markedly with pottery lipid extracts from all phases investigated from Bornais on the Outer Hebrides, where a partial or exclusive marine contribution was considerably less frequently identified even from contemporaneous pottery residues dating to the Mid- and Late Norse period. This lack of marine evidence on the sherds may reflect more intensive and specialized investment into marine resource exploitation by the 12th century on Shetland, and may be due to the types of fish being processed and the purpose for which these products were destined. We have also observed a strong emphasis on dairy products at Bornais. The residues from Jarlshof only exhibited limited evidence of dairy product processing, although the faunal remains are suggestive that a dairying strategy was also practiced at this settlement, and the Norse introduction of taxation in butter would suggest that dairying was likely important. However, since the majority of residues from Jarlshof pottery seem to have been used heavily for marine product processing, it is possible that milk products were processed in another kind of vessel not sampled. Ultimately, therefore, the question that remains is whether the organic residues actually reflect profound differences in the inhabitants’ diet at these two settlements or whether they result from different economic practices (e.g., production for export) or different product-processing techniques. Clearly, the data presented here represents only two sites from what is a likely to be a highly complex picture; however, it serves to highlight the differences in practices that may be investigated at other Norse settlements in Scotland. Acknowledgments We are grateful to Ian Bull, Alison Kuhl, and James Williams of the NERC Life Sciences Mass Spectrometry Facility for technical assistance and Marcus Badger for assistance producing the maps. This research was funded by Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 146 Copley, M.S., F. Hansel, K. Sadr, and R.P. Evershed. 2004. Organic residue evidence for the processing of marine animal products in pottery vessels from the pre-colonial archaeological site of Kasteelberg D East, South Africa. South African Journal of Science 100:279–283. Copley, M.S., H.A. Bland, P. Rose, M. Horton, and R.P. Evershed. 2005. Gas chromatographic, mass spectrometric, and stable carbon isotopic investigations of organic residues of plant oils and animal fats employed as illuminants in archaeological lamps from Egypt. Analyst 130:860–871. Cramp, L.J.E., and R.P. Evershed. 2014. Reconstructing aquatic resource exploitation in human prehistory using lipid biomarkers and stable isotopes. Pp. 319–339, In H.D. Holland and K.K. Turekian (Eds.).Treatise on Geochemistry, 2nd Edition. Vol 14. Elsevier, Oxford, UK. Crawford, I.A. 1981. War or Peace: Viking colonisation in the Northern and Western Isles of Scotland reviewed. Pp. 259–269, In H. Bekker-Nielson, P. Foote, and O. Olsen (Eds.). Proceedings of the Eighth Viking Congress, Aarhus, 24–31 August 1977. Odense University Press, Odense, Denmark. Dudd, S.N., and R.P. Evershed. 1998. Direct demonstration of milk as an element of archaeological economies. Science 282:1478–1481. Dudd, S.N., M. Regert, and R.P. Evershed. 1998. Assessing microbial contributions to absorbed acyl lipids during laboratory degradations of fats, oils, and pure triacylglycerols absorbed in ceramic potsherds. Organic Geochemistry 29:1345–1354. Dudd, S.N., R.P. Evershed, and A.M. Gibson. 1999. Evidence for varying patterns of exploitation of animal products in different prehistoric pottery traditions based on lipids preserved in surface and absorbed residues. Journal of Archaeological Science 26:1473–1482. Dunne, J., R.P. Evershed, M. Salque, L.J. Cramp, S. Bruni, K.Ryan, and S. di Lernia. 2012. First Dairying in “Green” Saharan Africa in the 5th Millennium BC. Nature 486:390–394. Evershed, R. P. 1993. Biomolecular archaeology and lipids. World Archaeology 25:74–93. Evershed, R.P. 2008. Experimental approaches to the interpretation of absorbed residues in archaeological ceramics. World Archaeology 40(1):26–47. Evershed, R.P., S. Vaughan, S.N. Dudd, and J. Soles. 1997. Fuel for thought? Beeswax in lamps and conical cups from Late Minoan Crete. Antiquity 71:979–985. Evershed, R.P., S.N. Dudd, M.S. Copley, and A.J. Mukherjee. 2002. Identification of animal fats via compound-specific δ13C values of individual fatty acids: Assessments of results for reference fats and lipid extracts of archaeological pottery vessels. Documenta Praehistorica 29:73–96. Evershed, R.P., M.S. Copley, L. Dickson, and F.A. Hansel. 2008. Experimental evidence for the processing of marine animal products and other commodities containing polyunsaturated fatty acids in pottery vessels. Archaeometry 50:101–113. Farquhar, G.D., J.R. Ehleringer, and K.T. Hubick 1989. Carbon isotope discrimination and photosynthesis. Annual Review of Plant Physiology and Plant Molecular Biology 40:503–537. NERC (NE/F021054/1). We would also like to thank Alison Sheridan of National Museums, Scotland, for providing archaeological material for sampling from Shetland; NERC for partial funding of the mass spectrometry facilities at Bristol (contract no. R8/H10/63; www.lsmsf.co.uk); Helen Grant of the NERC Life Sciences Mass Spectrometry Facility (Lancaster node) for stable isotopic characterization of reference standards and derivatizing; and the two anonymous reviewers for their helpful comments. Literature Cited Barrett, J.H. 1997. Fish trade in Norse Orkney and Caithness: A zooarchaeological approach. Antiquity 71:616–38. Barrett, J.H., and M.P. Richards. 2004. Identity, gender, religion, and economy: New isotope and radiocarbon evidence for marine resource intensification in Early Historic Orkney, Scotland, UK. European Journal of Archaeology 7:249–271. Barrett, J.H., R.A. Nicholson, and R. Cerón-Carrasco. 1999. Archaeo-ichthyological evidence for long-term socioeconomic trends in Northern Scotland: 2500–AD 1500. Journal of Archaeological Science 26:353–388. Barrett, J., R. Beukens, I. Simpson, P. Ashmore, S. Poaps, and J. Huntley. 2000. What was the Viking Age and when did it happen? A view from Orkney. Norwegian Archaeological Review 33(1):1–39. Barrett, J.H., R.P. Beukens, and R.A. Nicholson. 2001. Diet and ethnicity during the Viking colonisation of northern Scotland: Evidence from fish bones and stable carbon isotopes. Antiquity 75:145–154. Bigelow, G.F. 1985. Sandwick, Unst, and Late Norse Shetland economy. Pp. 95–127, In B. Smith (Ed.), Shetland Archaeology. The Shetland Times Ltd, Lerwick, UK. Cerón-Carrasco, R. 2005. “Of fish and men”: A study of the utilization of marine resources as recovered from selected Hebridean archaeological sites. Archaeopress (British Archaeological Reports British Series 400), Oxford, UK. Charters, S., R.P. Evershed, L.J. Goad, A. Leydon, P.W. Blinkhorn, and V. Denham. 1993. Quantification and distribution of lipid in archaeological ceramics: Implications for sampling potsherds for organic residue analysis and the classification of vessel use. Archaeometry 35(2):211–223. Charters, S., R.P. Evershed, A. Quye, P.W. Blinkhorn, and V. Reeves. 1997. Simulation experiments for determining the use of ancient pottery vessels: The behaviour of epicuticular leaf wax during boiling of a leafy vegetable. Journal of Archaeological Science 24(1):1–7. Christie, W.W. 1978. The composition, structure and function of lipids in the tissues of ruminant animals. Progress in Lipid Research 17: 111–205. Christie, W.W. 1981. Effects of diet on lipid composition. Pp. 203–226, In W.W. Christie (Ed.). Lipid Metabolism in Ruminant Animals. Pergamon Press, Oxford, UK. Copley, M.S., R. Berstan, S.N. Dudd, G. Docherty, A. Mukherjee, V. Straker, S. Payne, and R.P. Evershed. 2003. Direct chemical evidence for widespread dairying in prehistoric Britain. Proceedings of the National Academy of Sciences 100:1524–1529. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 147 Mottram, H.R., S.N. Dudd, G.J. Lawrence, A.W. Stott, and R.P. Evershed. 1999. New chromatographic, mass spectrometric, and stable isotope approaches to the classification of degraded animal fats preserved in archaeological pottery. Journal of Chromatography A 833:209–221. Mulville, J. 1999. The faunal remains. In N. Sharples (Ed.). The Iron Age and Norse Settlement at Bornais, South Uist. An Interim Report on the 1999 Excavations. Cardiff Studies in Archaeology Specialist Report 16, School of History and Archaeology, Cardiff University, Cardiff, UK. Available online at http:// www.cf.ac.uk/hisar/archaeology/reports/hebrides99/. Accessed 4 August 2011. Nicholson, R.A. 1998. Fishing in the Northern Isles: A case study based on fish bone assemblages from two multi-period sites on Sanday, Orkney. Environmental Archaeology 2:15–28. Regert, M., H.A. Bland, S.N. Dudd, P.F. van Bergen, and R.P. Evershed. 1998. Free and bound fatty acid oxidation products in archaeological ceramic vessels. Proceedings of the Royal Society of London B 265:2027–2032. Rieley, G. 1994. Derivatization of organic compounds prior to gas-chromatographic combustion-isotope ratio mass-spectrometric analysis: Identification of isotope fractionation processes. Analyst 119:915–919. Ritchie, A. 1974. Pict and Norsemen in northern Scotland. Scottish Archaeological Forum 6:23–36. Ritchie, A. 1977. Excavation of Pictish and Viking Age farmsteads at Buckquoy, Orkney. Proceedings of the Society of Antiquaries of Scotland 108:174–227. Sharples, N. 1999. The Iron Age and Norse settlement at Bornais, South Uist. An interim report on the 1999 Excavations. Cardiff Studies in Archaeology Specialist Report 16, School of History and Archaeology, Cardiff University, Cardiff, UK. Sharples, N. 2000. The Norse settlement at Bornais, South Uist. An interim report on the 2000 excavations. Cardiff Studies in Archaeology Specialist Report 18, School of History and Archaeology, Cardiff University, Cardiff, UK. Sharples, N. 2005. A Norse Farmstead in the Outer Hebrides. Excavations at Mound 3, Bornais, South Uist. Oxbow Books, Oxford, UK. Sharples, N., and M. Parker Pearson. 1999. Norse settlement in the Outer Hebrides. Norwegian Archaeological Review 32:41–62. Sharples, N., and R. Smith. 2009. Norse settlement in the Western Isles. Pp. 103–113, In A. Woolf (Ed.). Scandinavian Scotland: Twenty Years After. University of St. Andrews (St. John’s House Papers no. 12), St. Andrews, UK. Small, A. 1966. Excavations at Underhoull, Unst, Shetland. Proceedings of the Society of Antiquaries of Scotland 98:225–224. Smith, H., and J. Mulville. 2004. Resource management in the Outer Hebrides: An assessment of the faunal and floral evidence from archaeological investigations. Pp. 48–65, In R.A. Housley and G. Coles (Ed.). Atlantic Connections and Adaptations. Economies, Environments, and Subsistence in Lands Bordering the North Atlantic. Oxbow Books, Oxford, UK. Friedli, H., H. Lotscher, H. Oeschger, U. Siegenthaler, and B. Stauffer. 1986. Ice Core Record of the C-13/C-12 Ratio of Atmospheric CO2 in the Past 2 Centuries. Nature 324:237–238. Graham-Campbell, J., and C.E. Batey. 1998. Vikings in Scotland. An Archaeological Survey. Edinburgh University Press, Edinburgh, UK. Hamilton, J.R.C. 1956. Excavations at Jarlshof, Shetland. Archaeological Reports No. 1. Her Majesty’s Stationary Office, Edinburgh, UK. Hansel, F.A., M.S. Copley, L.A.S. Madureira, and R.P. Evershed. 2004. Thermally produced omega-(oalkylphenyl) alkanoic acids provide evidence for the processing of marine products in archaeological pottery vessels. Tetrahedron Letters 45(14):2999–3002. Hansel, F.A., and R.P. Evershed. 2009. Formation of dihydroxy acids from Z-monounsaturated alkenoic acids and their use as biomarkers for the processing of marine commodities in archaeological pottery vessels. Tetrahedron Letters 50(40):5562–5564. Hansel, F.A., I.D. Bull, and R.P. Evershed. 2011. Gas chromatographic mass spectrometric detection of dihydroxy fatty acids preserved in the “bound” phase of organic residues of archaeological pottery vessels. Rapid Communications in Mass Spectrometry 25:1893–1898. Heaton, T.H.E. 1999. Spatial, species, and temporal variations in the 13C/12C ratios of C3 plants: Implications for palaeodiet studies. Journal of Archaeological Science 26:637–649. Heron, C., and R.P. Evershed. 1991. Effects of migration of soil lipids on organic residues associated with buried potsherds. Journal of Archaeological Science 18:641–659. Ingrem, C. 2000. The fish bones. In N. Sharples (Ed.). The Norse Settlement at Bornais, South Uist. An Interim Report on the 2000 Excavations. Cardiff Studies in Archaeology Specialist Report 18, School of History and Archaeology, Cardiff University, Cardiff, UK. Available online at http://www.cf.ac.uk/hisar/archaeology/ reports/hebrides00/. Accessed 4 August 2011. Ingrem, C. 2005. Site activities: Fish. Pp. 176, In N. Sharples (Ed.). A Norse Farmstead in the Outer Hebrides. Excavations at Mound 3, Bornais, South Uist. Oxbow Books, Oxford, UK. Kolattukudy, P.E., R. Croteau, and J.S. Buckner. 1976. Biochemistry of plant waxes. Pp. 290–327, In P.E. Kolattukudy (Ed.). Chemistry and Biochemistry of Natural Waxes. Elsevier, Amsterdam, The Netherlands. Matikainen, J., S. Kaltia, M. Ala-Peijari, N. Petit-Gras, K. Harju, J. Keikkila, R. Yksjarvi, and T. Hase. 2003. A study of 1,5 hydrogen shift and cyclisation reactions of isomerised methyl linoleate. Tetrahedron Letters 59:567–573. McGregor, A. 1880. Notes on some old customs in the Isle of Skye. Proceedings of the Society of Antiquaries of Scotland 14:143–147. Morris, C.D. 1996. The Norse impact in the Northern Isles of Scotland, Pp. 69–83, In J.F. Kroger and H.-R. Naley (Eds.). Nordsjoen: Handel, Religion, og Politick. Dreyer Bok, Stavanger, Norway. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 148 Tulloch, A.P. 1976. Chemistry of waxes of higher plants. Pp. 236–287, In P.E. Kolattukudy (Ed.). Chemistry and Biochemistry of Natural Waxes. Elsevier, Amsterdam, The Netherlands. Vetter, W., and M. Schröder. 2011. Phytanic acid: A tetramethyl- branched fatty acid in food. Lipid Technology 23:185–178. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 149 Appendix 1. Methodology. Extraction and derivatization of total lipid extracts Any carbonized deposits visible on the inner or outer surfaces of the pot were first removed using a solvent-cleaned scalpel, to be investigated alongside lipids extracted from the fabric of the pottery. The outer surfaces of a small area of the pot sherds were then cleaned using a modelling drill fitted with an abrasive drill bit, to remove surface contamination derived from the burial surroundings or lipids inadvertently introduced through post-excavational handling. Approximately 2 g cleaned sherd was removed with a chisel, and these sherds and portions of the carbonized residues were ground to a fine powder using a mortar and pestle. Immediately prior to lipid extraction, 20 μg or 10 μg of a C34 n-alkane was added to the crushed pot or carbonized residue, respectively, as an internal standard, and lipids were then extracted (2 x 10 ml CHCl3/ MeOH 2:1 v/v; sonication x 20 min). Aliquots of the resulting total lipid extract (TLE) were filtered through a silica column and treated using 40 μl N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA; 70 °C, 1 h). Excess BSTFA was removed under a gentle flow of nitrogen prior to dissolution of the residue in an appropriate volume of hexane. Extraction and derivatisation of “bound” lipids Base extractions to release the “bound” lipids were performed on ≈1 g previously solvent-extracted pottery. Five millilitres 0.5M NaOH/MeOH was added, and the mixture heated (70 °C, 1 h). After cooling to room temperature, the mixture was acidified to pH 3 using 1 M aqueous HCl, and then the lipids extracted using 3 x 3 ml CHCl3. This fraction was analyzed as TMS derivatives as described above. Preparation of FAMEs For the detection of isoprenoid and ω(o-alkylphenyl)alkanoic acids and the determination of the stable carbon isotope composition of individual fatty acids, aliquots of selected TLEs were prepared for analysis via GC/MS, GC/MS-SIM, and GC/C/IRMS as fatty acid methyl esters (FAMEs). Portions of lipid extracts were hydrolyzed using 2 ml 0.5 M NaOH/MeOH (70 °C, 1 h). The mixture was cooled, and the neutral fraction extracted using 3 x 3 ml hexane. The sample was acidified to pH 3 using 1 M aqueous HCl and the lipids extracted using 3 x 3 ml CHCl3. The hydrolyzed fatty acids were then methylated using 100 μl BF3/MeOH (14% w/v; 75 °C, 1 h), and the FAMEs extracted using 3 x 2 ml CHCl 3. High temperature gas chromatography (HTGC) All archaeological total and “bound” lipid extracts were screened and quantified using an Agilent 7890 fitted with a high-temperature non-polar column with a 100% dimethyl polysiloxane stationary phase (15 m x 0.32 mm i.d., 0.1 μm film thickness). The temperature program comprised a 50 °C isothermal followed by an increase to 350 °C at a rate of 10 °C min-1 followed by a 10-min isothermal. Gas chromatography/mass spectrometry (GC/MS) Base extracts were analyzed using a GC/MS fitted with a non-polar column for the identification of hydroxy fatty acids. The instrument was a ThermoFinnigan single quadrupole TraceMS run in electron ionisation (EI) mode (electron energy 70 eV, scan range of m/z 50–650 and scan time of 0.6 s). Samples were introduced onto a non-polar column with a 100% dimethyl polysiloxane stationary phase (CP-Sil CB, 50 m x 0.32 mm i.d., 0.12-μm film thickness) with an oven program comprising a 2-min isothermal at 50 °C, increasing to 300 °C at a rate of 10 °C min-1, followed by a 10-min isothermal. FAMEs were analyzed on a ThermoFinnegan single quadrupole TraceMS run in EI mode (electron energy 70 eV). The oven program comprised a 2-min isothermal hold at 50 °C, increasing at 10 °C min-1 to 100°C and then ramping at 4 °C min-1 to 240 °C, ending with a 15-min isothermal. Samples run in full-scan mode had a scan range of m/z 50–650 and scan time of 0.6 s. The GC/MS was fitted with a polar column (VF23 MS, high cyano-modified cyanopropyl polysilphenylenesiloxane, 60 m x 0.32 mm i.d., 0.15 μm film thickness). Extracts analyzed in selected ion monitoring mode followed an identical temperature program, and the MS was programmed to scan for ions m/z 105, 262, 290, 318, and 346. GC/C/IRMS GC-C-IRMS analyses were performed using an Agilent 6890 GC coupled to a Deltaplus XL via a Finnegan MAT GC combustion III interface. The GC was fitted with a non-polar (CP-Sil CB, 100% dimethylpolysiloxane, 50 m x 0.32 i.d., 0.12 μm film thickness) column. Samples were introduced via a split/splitless injector in splitless mode. The temperature program comprised a 2-min isothermal at 50 °C followed by an increase of 10 °C min-1 to 300 °C and a 10-min isothermal. The injector temperature started at 70 °C and increased to 300 °C at a rate of 600 °C min-1 and was then held for 30 min. Faraday cups were used to detect ions of m/z 44 (12C16O2), 45 (13C16O2 and 12C17O16O), and 46 (12C18O16O). Results were calibrated against reference CO2 that was injected directly into the source 3x at the beginning and end of the run. All samples were run in duplicate, and any runs with inacceptable reproducibility, standard values, peak shapes, or peak intensities were discarded and repeated. External standards were run every 4 runs. The δ13C values were derived according to the following expression and are relative to the international standard vPDB: δ13C‰ = ([R sample - R standard] / R standard) x 1000, where R = 13C/12C. The δ13C values were corrected for the carbon atoms added during methylation using a mass balance equation (Rieley 1994). Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 150 Appendix 2. Table showing the stable carbon isotope values and the presence of marine-derived biomarkers in sherds containing significant quantities of lipid. APAAs: ω-(o-alkylphenyl)alkanoic acid; diOH FAs: dihydroxy fatty acid. D (x) = detected (no.); “-” = not detected; n.a. = not analyzed. Sherd House δ13C16:0‰ δ13C18:0‰ Δ13C‰ APAAs diOH FAs Isoprenoid FAs JAR-1 7 -22.2 -21.8 0.4 C18-C22 n.a. n.a. JAR-2 7 -22.2 -22.5 -0.2 C18-C22 n.a. n.a. JAR-3 7 -22.2 -25.6 -3.1 C18-C22 n.a. n.a. JAR-5 7 -27.5 -30.5 -3.0 C18 n.a. n.a. JAR-6 7 -23.2 -25.3 -2.1 C18-C22 C18-C22 D (3) JAR-7 7 -23.6 -24.7 -1.1 C18-C22 C18-C22 n.a. JAR-8 7 -22.7 -22.4 0.3 C18-C22 C18-C22 n.a. JAR-12 7 -26.2 -28.1 -1.9 - n.a. n.a. JAR-13 7 -22.6 -24.3 -1.7 C18-C22 C18-C22 D (2) JAR-14 5 -26.1 -32.7 -6.6 C18-C22 C18-C22 n.a. JAR-16 7 -27.4 -30.1 -2.7 C18-C20 n.a. n.a. JAR-17 7 -25.2 -29.0 -3.8 - n.a. n.a. JAR-18 7 -21.6 -23.6 -2.0 C18-C22 n.a. n.a. JAR-19 5 -23.8 -24.0 -0.2 C18-C22 C18-C22 - JAR-21 5 -21.7 -21.4 -0.3 C18 n.a. n.a. JAR-22 5 -24.1 -27.0 -2.9 C18-C22 C18-C22 D (1) JAR-23 - -25.0 -26.3 -1.3 C16-C22 n.a. D (3) JAR-24 - -26.2 -29.5 -3.3 C16-C22 n.a. D (1) JAR-25 - -24.8 -27.5 -2.7 C16-C22 n.a. D (3) JAR-14V 5 -24.7 -31.7 -7.0 C18-C22 n.a. - JAR-15V 5 -22.2 -21.2 1.0 C18-C22 n.a. n.a. JAR-22V 5 -25.6 -29.2 -3.6 C18-C22 n.a. n.a. BN-17 LIA 1 -27.0 -30.2 -3.2 C18 - D (2) BN-19 LIA 1 -26.9 -30.8 -3.9 C18 n.a - BN-20 LIA 1 -27.1 -31.0 -3.9 - n.a - BN-25 LIA 1 -27.6 -32.5 -4.8 C18 - - BN-58 LIA 1 -28.3 -32.0 -3.7 - - n.a. BN-62 LIA 1 -27.6 -32.5 -4.8 - n.a n.a. BN-66 LIA 2 -27.4 -30.2 -2.9 C18 C18 n.a. BN-66V LIA 2 -27.4 -30.2 -2.9 C18 n.a n.a. BN-68 LIA 2 -26.8 -28.5 -1.6 - n.a n.a. BN-69 LIA 2 -26.9 -31.0 -4.0 C18-C20 C18-C22 n.a. BN-70 LIA 2 -26.8 -29.9 -3.1 C18-C20 C18 n.a. BN-70V LIA 2 -26.8 -29.3 -2.5 C18 n.a n.a. BN-71 LIA 2 -27.4 -31.7 -4.3 C18 C18 n.a. BN-71V LIA 2 -26.9 -30.8 -3.8 C18 n.a n.a. BN-72 LIA2 -28.1 -33.8 -5.8 - - n.a. BN-74 LIA2 -26.6 -29.4 -2.8 C18-C20 C18 n.a. BN-74V LIA2 -26.9 -29.8 -2.9 C18 n.a n.a. BN-76 LIA2 -26.8 -29.3 -2.5 C18 C18 n.a. BN-77 LIA2 -26.9 -29.9 -3.0 C18 C18 n.a. BN-78V LIA2 -27.0 -29.7 -2.7 n.a n.a n.a. BN-79 LIA2 -26.8 -30.0 -3.2 n.a n.a n.a. BN-80 LIA2 -28.3 -32.5 -4.3 n.a - n.a. BN-82 LIA2 -26.8 -29.0 -2.1 n.a n.a n.a. BN-83 LIA2 -27.4 -30.6 -3.3 C18 C18-C22 n.a. BN-84 LIA2 -28.0 -30.4 -2.4 - n.a n.a. BN-86 LIA2 -26.8 -29.4 -2.5 - n.a n.a. BN-87 LIA2 -26.8 -30.4 -3.6 C18-C20 C18 n.a. BN-87V LIA2 -27.5 -30.7 -3.2 - n.a n.a. BN-88 LIA2 -24.2 -25.4 -1.3 C18-C22 C18-C22 - BN-88V LIA2 -25.1 -28.5 -3.4 - n.a n.a. BN-89 LIA2 -26.8 -29.4 -2.5 C18 n.a n.a. BN-35 E Norse -25.4 -31.4 -6.0 C18-C20 C18-C22 D (1) BN-35V E Norse n.a n.a n.a C18-C20 n.a - BN-90 E Norse -26.4 -29.9 -3.6 - n.a n.a. Journal of the North Atlantic L.J.E. Cramp, H.Whelton, N. Sharples, J. Mulville, and R.P. Evershed 2015 Special Volume 9 151 Sherd House δ13C16:0‰ δ13C18:0‰ Δ13C‰ APAAs diOH FAs Isoprenoid FAs BN-91 E Norse -24.2 -27.9 -3.6 C18-C20 C18-C22 n.a. BN-92 E Norse -27.6 -30.9 -3.4 - n.a n.a. BN-93 E Norse -26.4 -30.9 -4.6 C18-C20 C18 n.a. BN-94 E Norse -26.5 -28.7 -2.3 - n.a n.a. BN-95 E Norse -26.6 -29.3 -2.7 C18 n.a n.a. BN-96V E Norse -26.9 -31.1 -4.3 - n.a n.a. BN-97 E Norse -27.4 -30.3 -2.9 - n.a n.a. BN-99 E Norse -29.01 -30.5 -1.4 - n.a n.a. BN-100 E Norse -28.9 -30.5 -1.6 - n.a n.a. BN-101 E Norse -28.1 -32.2 -4.1 C18 n.a n.a. BN-103 E Norse -28.0 -30.1 -2.1 - n.a n.a. BN-104 E Norse -27.3 -30.9 -3.7 - n.a n.a. BN-105 E Norse -26.2 -31.7 -5.4 C18-C22 C18-C22 - BN-107 E Norse -25.9 -32.3 -6.4 - n.a n.a. BN-108 E Norse -27.5 -32.7 -5.2 C18 n.a n.a. BN-110 E Norse -25.4 -25.9 -0.5 C18-C20 C18 n.a. BN-111V E Norse -26.8 -32.4 -5.6 C18 n.a n.a. BN-28V M Norse n.a n.a n.a C18-C20 n.a D (3) BN-29 M Norse -27.0 -31.3 -4.3 - n.a D (1) BN-34 M Norse -26.3 -30.5 -4.2 C18 - D (1) BN-34V M Norse -25.0 -28.7 -3.6 C18-C20 n.a D (1) BN-45 M Norse n.a n.a n.a - n.a - BN-115 M Norse -27.2 -31.0 -3.8 C18 n.a n.a. BN-115V M Norse -27.8 -30.6 -2.8 - n.a n.a. BN-117 M Norse -27.3 -30.1 -2.8 - n.a n.a. BN-118 M Norse -26.7 -30.1 -3.4 C18 n.a n.a. BN-121 M Norse -26.8 -32.0 -5.2 - n.a n.a. BN-122 M Norse -28.0 -30.8 -2.8 - n.a n.a. BN-123 M Norse -26.5 -31.8 -5.4 C18 n.a - BN-123V M Norse -26.9 -31.4 -4.5 C18 n.a n.a. BN-124 M Norse -27.2 -29.8 -2.6 C18 n.a n.a. BN-124V M Norse -27.3 -29.7 -2.4 C18 n.a n.a. BN-33 M-L Norse -27.6 -33.6 -6.0 - n.a - BN-38 M-L Norse -25.7 -29.4 -3.7 C18-C20 C16-C18 D_(1) BN-51 M-L Norse -26.4 -28.9 -2.5 - - - BN-53 M-L Norse -26.4 -29.4 -3.0 C18-C20 n.a D (1) BN-36 L Norse -27.3 -31.8 -4.5 C18-C20 C16-C22 - BN-39 L Norse -27.6 -31.8 -4.2 C18-C20 - D (1) BN-39V L Norse n.a n.a n.a - n.a n.a BN-44 L Norse n.a n.a n.a - n.a - BN-46 L Norse n.a n.a n.a - n.a - BN-46V L Norse n.a n.a n.a - n.a n.a BN-48 L Norse n.a n.a n.a - n.a - BN-49 L Norse -27.2 -31.6 -4.5 C18 - - BN-52 L Norse -26.5 -29.1 -2.6 C18-C20 n.a D (2) BN-55 L Norse n.a n.a n.a C18-C20 C16-C20 n.a. BN-125 L Norse -28.2 -30.7 -2.5 C18 n.a n.a. BN-126 L Norse -28.2 -32.5 -4.4 C18 n.a n.a. BN-126V L Norse -27.1 -32.4 -5.3 C18 n.a n.a. BN-130 L Norse -28.3 -33.0 -4.8 - n.a n.a. BN-131 L Norse -26.8 -30.8 -4.0 C18 n.a - BN-8 Norse -27.6 -33.4 -5.8 - - D (1) BN-8V Norse n.a n.a n.a C18 n.a - BN-47 Norse -28.4 -33.6 -5.1 n.a n.a -