Abstract
We investigated the contribution of freshwater resources to the diet of seven Late Mesolithic hunter-gatherers (ca. 5300–7000 BC) from Northern France and Luxembourg using stable isotope ratios. In addition to the carbon and nitrogen stable isotope ratios (δ13C, δ15N), we explored the potential of the sulphur isotopic ratios (δ34S) to detect and quantify the proportion of protein derived from aquatic foodstuff. In only two sites, animal remains from an associated settlement were available and subsequently examined to decipher the isotopic differential between terrestrial and freshwater resources. The quantification of their relative contribution was simulated using a Bayesian mixing model. The measurements revealed a significant overlap in δ13C values between freshwater and terrestrial resources and a large range of δ15N values for each food category. The δ34S values of the aquatic and terrestrial animals were clearly distinct at the settlement in the Seine valley, while the results on fish from Belgium demonstrated a possible overlap in δ34S values between freshwater and terrestrial resources. Local freshwater ecosystem likely contributed to ca. 30–40 % of the protein in the diet of the individuals found in the Seine settlement. Out of this context, the isotopic signature and thus contribution of the available aquatic foods was difficult to assess. Another potential source of dietary protein is wild boar. Depending on the local context, collagen δ34S values may contribute to better assessment of the relative contribution of freshwater and terrestrial resources.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Avoid common mistakes on your manuscript.
Introduction
For the last few decades, the application of stable isotope ratio analyses on Late Mesolithic hunter-gatherers in Europe has mainly been considered in the context of the transition to the first farming communities of the Neolithic and their relative use of aquatic resources in this important transitional period (e.g. Lubell et al. 1994; Richards and Mellars 1998; Lillie and Richards 2000; Richards et al. 2003a, c; Bonsall et al. 2004; Bocherens et al. 2007; Fischer et al. 2007; Smits and van der Plicht 2009; Lightfoot et al. 2011; Schulting and Richards 2001; Guiry et al. 2015). Most of the studies quoted above were conducted on sites from coastal environments where the consumption of marine resources was investigated. To date, research on the use of freshwater resources by Mesolithic groups in a continental context has been restricted to relatively few specific geographical areas such as the Danube Gorges (e.g. Bonsall et al. 2004; Borić et al. 2004; Nehlich et al. 2010), the Dnieper Basin (Lillie et al. 2011), the Meuse Basin (Bocherens et al. 2007) and western Germany (Bollongino et al. 2013).
An increasing number of Late Mesolithic burials have recently been added to the archaeological record of continental Northwestern Europe, notably in Northern France (e.g. Valentin et al. 2008; Meiklejohn et al. 2010; Bosset and Valentin 2013). It contrasts with the poorly represented Late Mesolithic archaeological contexts of Belgium and the yet sole example of Loschbour in Luxembourg (review in Meiklejohn et al. 2014). A large number of the Northern France burials are associated with very few, if any, artefacts and are not associated to settlement context. Based on faunal remains found in Northern France and Luxembourg, animal food resources appear to include the wild boar (Sus scrofa), red deer (Cervus elaphus), roe deer (Capreolus capreolus) and aurochs (Bos primigenius) (e.g. Cordy 1982; Bridault 1997; Ducrocq et al. 2008; Leduc et al. 2013; Marinval-Vigne et al. 1989). Evidence for fishing is limited until 8200 14C BP and confidently attested at only two sites: Noyen-sur-Seine and La Chaussée-Tirancourt in Northern France (Marinval-Vigne et al. 1989; Ducrocq and Ketterer 1995).
Stable isotope ratios and paleodiet
Carbon and nitrogen isotope abundances in bone collagen have proved to be useful tracers of the source of protein, namely terrestrial versus aquatic, in the diet of ancient hunter-gatherers in continental context (e.g. Richards et al. 2001; Bocherens et al. 2007; Drucker et al. 2016). For predators, the collagen δ13C and δ15N values depend on those of their food with a factor of enrichment, which is limited in 13C but significant in 15N (0.8 to 1.3 ‰ compared with 3 to 5 ‰ in the latter; e.g. Bocherens and Drucker 2003). For a comparable trophic position, specimens from aquatic ecosystems deliver higher 15N abundances than specimens from terrestrial context (e.g. Schoeninger and DeNiro 1984; Dufour et al. 1999), a difference that is reflected in their consumers. However, the respective ranges of δ15N values of terrestrial and freshwater resources also depend on environmental factors such as temperature and aridity that may differ significantly geographically and temporally (e.g. Drucker et al. 2003; Bocherens et al. 2014). This can result in a large range of isotopic variation and thus overlapping isotopic values between potential prey species (e.g. Dufour et al. 1999; Katzenberg et al. 2010).
The 34S abundances in collagen have increasingly been investigated on archaeological remains to track aquatic resources consumption (review in Nehlich 2015). Primary producers are characterized by δ34S values around +20 ‰ in oceanic contexts, whereas, in continental ecosystems, they exhibit lower δ34S values except in areas close to coastlines due to the sea spray effect (e.g. Peterson and Fry 1987). The δ34S values of plants are passed on along the food chain with minor fractionation from ca. −1 to +1.5 ‰ (Richards et al. 2003b; Barnes and Jennings 2007; Tanz and Schmidt 2010) if not negligible (McCutchan et al. 2003; Arneson and MacAvoy 2005). Depleted δ34S values are observed as the result of the reduction of sulphate (SO4 2−) to hydrogen sulphide (H2S) and subsequent reoxidation by bacteria, which are notably present in rivers and aquatic sediments (Fry et al. 1986; Holmer and Storkholm 2001). Hence, a distinction in the abundance of 34S is expected between terrestrial and freshwater ecosystems and has indeed already been explored in ancient contexts (Privat et al. 2007; Nehlich et al. 2010, 2011; Bocherens et al. 2011). Although a clear distinction in the collagen δ34S values was found in most cases between terrestrial and freshwater faunal remains, the observed ranges of values were variable from one case study to the other due to local differences in geology, atmospheric deposition and conditions of sulphur cycling. The pattern of distinction in 34S abundances between terrestrial and freshwater resources is thus unpredictable, and systematic analyses of the local faunal remains are required to establish it.
In this paper, we aim to reconstruct the contribution of freshwater resources to the diet of Late Mesolithic humans from Northern France and Luxembourg (ca. 6300–8000 14C BP, ca. 5300–7000 BC) using 13C and 15N abundances of their bone collagen. For this purpose, we established the isotopic baseline using as many animal remains as possible from the same region and time span to decipher the specific stable isotope pattern of the associated ecosystems. We considered, in addition, the abundances in 34S in order to explore further quantitative estimation of the different sources of protein, in particular between freshwater and terrestrial and possibly fine-tune the reconstruction of subsistence at an individual level.
Materials and methods
Materials
The studied human remains come from different sites in Northern France and Luxembourg (Fig. 1) and witnessed diverse funerary treatments. The site of Loschbour at Heffingen in the Grand-Duchy of Luxembourg, discovered in 1935, provided a primary inhumation (LSB1) with a cremation (LSB2) (Toussaint et al. 2009). The excavation also revealed evidence of a settlement area with a lithic industry of Montbanian culture and some faunal remains predominantly of not only wild boar but also aurochs, red deer, roe deer and beaver (Gob 1982). The buried individual was a male adult whose skull was partially ochered (Delsate et al. 2009, 2011). Two aurochs ribs in apparent association with the human individual were dated to 7115 ± 45 14C BP (Gob 1982).
Geographical location of the sites investigated in this study
Regarding Northern France, Maisons-Alfort at Zac d’Alfort (Val de Marne, France) corresponds to a primary single burial. The individual is a mature adult of unknown sex due to the poor preservation of the remains (Valentin et al. 2008). Similarly, the sites of Cuiry-lès-Chaudardes ‘Les Fontinettes’ and Berry-au-Bac ‘Le Vieux Tordoir’ have both provided a single human burial with a necklace made of pike vertebrae in the first case (Ilett 1998) and with ochre and a bone artefact in the latter. In contrast, the human remains at Noyen-sur-Seine ‘Hauts des Nachères’ were found in a non-funerary context. The site is a riverbank formation located between Nogent-sur-Seine and Montereau. Archaeological remains from the four excavated loci reveal a Mesolithic occupation in palaeochannels of the Seine River in a peat context. Surface erosion led to the exposure of the peat deposit 9sup in locus 3, while the lower peat deposit 9 is visible in the excavated locus 2 (reviewed in Mordant et al. 2013). One radiocarbon date was obtained on the wood fibres of a fish trap (8000 ± 100 14C BP) recovered at the basis of layer 9. Two more recent dates were retrieved on wood fragments from level 9sup: 7040 ± 80 14C BP and 6240 ± 70 14C BP. A preliminary study of a sub-sample of the human remains, found in both level 9sup and level 9, described cranial and postcranial remains of varying degrees of completeness and, in some cases, showing superficial modifications such as traces of burning and cutmarks (Auboire 1991). In this paper, we investigate one adult and one juvenile individual from level 9 and one adult from level 9sup on which further study is currently conducted for age and sex determination. With the exception of the non-adult individual from layer 9 of Noyen-sur-Seine (NO7300), the individuals under consideration were directly radiocarbon dated (Table 1).
The interpretation of the isotopic signature of the human individuals requires the establishment of isotopic baseline provided by local faunal remains. The terrestrial animals we included are red deer (C. elaphus) and wild boar (S. scrofa) from both Noyen-sur-Seine and Loschbour. In addition, roe deer (Capreolus capreolus), aurochs (B. primigenius), wild boar (S. scrofa), wolf (Canis lupus), red fox (Vulpes vulpes), lynx (Lynx lynx), and wild cat (Felis silvestris) were sampled from the faunal assemblage of Noyen-sur-Seine. With the exception of one unfused long bone of aurochs (NO5200), only mature individuals were sampled from the animal remains and, as far as possible, the same anatomical part was selected. Aquatic animals chosen for sampling encompass European pond turtle (Emys orbicularis), northern pike (Esox lucius) and European eel (Anguilla anguilla) as well as one of their predator, the otter (Lutra lutra), all specimens coming from Noyen-sur-Seine. One pike and several cyprinids from the Mesolithic occupation of Abri du Pape located along the Meuse river in Belgium, with radiocarbon dates ranging from 8800 to 7850 14C BP (in Léotard et al. 1999) supplemented the dataset. Magdalenian fish freshwater species (ca. 12,800 14C BP) including burbot (Lota lota), brown trout (Salmo trutta fario), and nase (Chondrostoma nasus) from earlier Magdalenian sites of the Meuse valley (Drucker et al. 2016) were also added in the present analysis.
Sample preparation and analysis
Collagen was extracted following a protocol based on Longin (1971) and modified by Bocherens et al. (1997). The extraction process includes a step of soaking in 0.125 M NaOH between the demineralization and solubilization steps to achieve the elimination of lipids. Elemental analysis (C, N, S) and isotopic analysis (δ13C, δ15N, δ 34S) were conducted at the Department of Geosciences of Tübingen University using a NC2500 CHN-elemental analyser coupled to a Thermo Quest Delta + XL mass spectrometer. Sample δ13C and δ15N values are reported relative to the international reference scales V-PDB for carbon and AIR for nitrogen isotopes. Analytical error, based on within-run replicate measurement of laboratory standards (albumen, modern collagen, USGS 24, IAEA 305A), was ±0.1 ‰ for δ13C values and ±0.2 ‰ for δ15N values. Samples were calibrated to δ34S values relative to V-CDT of NBS 123 (δ34S = 17.1 ‰), NBS 127 (δ34S = 20.3 ‰), IAEA-S-1 (δ34S = −0.3 ‰) and IAEA-S-3 (δ34S = 21.7 ‰). The reproducibility is ±0.4 ‰ for δ34S measurements, and the error on S measurement is 5 %. Reliability of the δ13C and δ15N values can be established by measuring collagen chemical composition, with atomic C:N ranging from 2.9 to 3.6 (DeNiro 1985) and the percentage of C and N above 8 and 3 %, respectively (Ambrose 1990). We retained δ34S values of samples whose atomic C:S and N:S fit into the range of 300–900 and 100–300, respectively (Nehlich and Richards 2009) and whose percentage of S ranged between 0.14 and 0.26 %, determined through the results of modern mammalian collagen (Aldrich collagen, modern elk and modern camel) measured in the same sets. The 34S amounts and the δ34S values in animal species from level 9 of Noyen-sur-Seine published in Bocherens et al. (2011) were normalised based on Sigma collagen (0.22 ± 0.003 % and 4.2 ± 0.2 ‰) and Aldrich collagen (0.18 ± 0.01 % and 1.8 ± 0.6 ‰) measurements used in Drucker et al. (2011).
Direct AMS radiocarbon dates have been performed on most of the considered human remains and some faunal specimens (Table 1). The results were then calibrated with a 95.4 % confidence level and rounded to the nearest 5 based on the OxCal 4.2 programme (Bronk Ramsey and Lee 2013) using the IntCal13 calibration curve (Reimer et al. 2013).
Calculation of the proportions of consumed preys
The relative contribution of the different preys to the average diet of the human individuals was simulated using a Bayesian mixing model approach performed in the Stable Isotope Analysis in R (SIAR) package (Parnell et al. 2010), using the R software, version 3.0.2 (Team R Core 2013). SIAR offers the possibility to incorporate uncertainty in input data and yields not only a range of possible dietary proportions but also provides also their relative probability distribution (Parnell et al. 2010). We considered a trophic enrichment factor (TEF) of +1.1 ± 0.2 and +3.8 ± 1.1 ‰ for δ13C and δ15N values, respectively, based on a review comparing the δ13C and δ15N values of the bone collagen of predators and that of their prey in modern and ancient terrestrial ecosystems (reviewed in Bocherens et al. 2015). There are however few controlled feeding experiments on mammals to enlighten TEF between consumer body tissues and their diet for δ34S (Richards et al. 2003b; Tanz and Schmidt 2010). In this study, we used a TEF of 0.5 ± 2.4 ‰ for δ34S corresponding to an average of published TEF data for modern fauna (reviewed in Nehlich 2015).
Results and discussion
δ13C and δ15N values of the preys
The δ13C values of the collagen of the terrestrial fauna of the Mesolithic occupation of Noyen-sur-Seine and Loschbour revealed some inter-species differences (Table 2, Fig. 2). Among the terrestrial ungulates, the wild boar showed δ13C values varying from −21.5 to −18.8 ‰, while a range of lower values were observed for aurochs, red deer and roe deer (−23.8 to −22.3 ‰). The latter are similar to collagen values reported for herbivores consuming plants that developed under a dense canopy (see review in Drucker et al. 2008). Several studies in boreal and temperate forest confirmed the occurrence of a ‘canopy effect’ on understory plants (Broadmeadow et al. 1992; Brooks et al. 1997; Bonafini et al. 2013) due to the accumulation of 13C-depleted CO2 produced by leaf litter recycling and the change in photosynthetic activity and stomatal conductance linked to the specific conditions of light and humidity (e.g. Francey et al. 1985; van der Merwe and Medina 1991; Broadmeadow et al. 1992). Thus, a closed forest habitat can be deduced for the aurochs and roe deer of Noyen-sur-Seine and the red deer at both Noyen-sur-Seine and Loschbour as both show very similar δ13C values range (−23.1 to −22.3 and −23.1 to −22.5 ‰, respectively). In contrast, the relatively high δ13C values of the wild boar of Noyen-sur-Seine and Loschbour could be due to their specific consumption of fruits, acorns and underground tubers (e.g. Ballari and Barrios García 2014). No canopy effect is expected for these food items since they come from the top of the canopy or develop underground. It would be thus premature to conclude that the habitat of the wild boar was not overlapping that of the deer and large bovine. Hence, the temperate forest can be considered as the permanent habitat of a large variety of the terrestrial preys hunted by the Mesolithic groups around Noyen-sur-Seine and Loschbour.
δ13C and δ15N values of bone collagen of the red deer (n = 7), roe deer (n = 7), wild boar (n = 7), aurochs (n = 5), fish (n = 20), pond turtle (n = 6), wolf (n = 1), red fox (n = 1), lynx (n = 1), wild cat (n = 1) and otter (n = 3)
Most of the δ15N values of the ungulates clustered between 3.7 and 7.4 ‰ with comparable range from one species to another (Table 2, Fig. 2). However, two specimens from Noyen-sur-Seine, roe deer NO5500 and wild boar NO1100, showed a higher δ15N value of 10.0 and 9.2 ‰, respectively, and so did a wild boar sample from Loschbour as well. The high 15N abundance of the Loschbour specimen can be explained by the nursing effect since the collagen was extracted from a canine root. The enriched 15N collagen from the two samples of Noyen-sur-Seine could be explained by the consumption of a specific enriched 15N plant, such as mushrooms (e.g. Drucker et al. 2012), locally specific conditions of temperature and/or aridity (e.g. Amundson et al. 2003; Craine et al. 2009) or a forage quality significantly under the animal’s requirement (e.g. Robbins et al. 2010; Poupin et al. 2011). An alternative reason for the case of the wild boar could be the access to animal protein since this species can be omnivorous (e.g. Ballari and Barrios García 2014). Such a hypothesis was confirmed through the 15N analysis of single amino acids (Naito et al. 2013). Even though the number of these high 15N ungulates was limited, they might have introduced variability in the δ15N values of the terrestrial predators.
δ13C and δ15N values of the animal predators
Within the terrestrial carnivores, among which considerable variation in isotopic composition was observed, two groups can be considered on the basis of their δ15N values. The first group corresponded to a δ15N value of 9.3 ‰ and included the wild cat and the wolf (δ13C value of −19.8 and −21.3 ‰, respectively). The red fox and the lynx both yielded δ15N values around 12 ‰, but δ13C in the same range as the wild cat and the wolf. These last predators presented the expected enrichment of 3 to 5 ‰ in δ15N and about 1 ‰ in the δ13C (Bocherens and Drucker 2003) in relation to the averaged values of most of the terrestrial ungulates. The more enriched 15N predators can be explained by a consumption of ungulates with high δ15N values such as the wild boar NO1100 and the roe deer NO5500 or young animals since they constitute a source of 15N-enriched meat due to the nursing effect. An opportunistic behaviour can be hypothesised for the red fox, suggesting the scavenging of dietary remains of human groups, including larger preys than those normally hunted by it (Drucker 2001) or some aquatic resources in this context. Indeed, the consumption of large ungulates and aquatic animals through scavenging has been documented for red fox in modern temperate ecosystems (e.g. Jędrzejewski and Jędrzejewska 1992; Lanszki 2005). In conclusion, the contrasted isotopic signatures of the terrestrial carnivores reflect the isotopic diversity of the accessible preys.
The Mesolithic fish from Noyen-sur-Seine and the Abri du Pape clustered between −23.7 and −21.2 ‰ for the δ13C values and between 8.0 and 10.9‰ for the δ15N values (Table 2, Fig. 2). These values fall within the wide range found for fish of the Belgian Magdalenian (−24.1 to −18.7 ‰ in δ13C, 6.6 to 11.7 ‰ in δ15N; Drucker et al. 2016). While the δ13C values of the fish largely overlapped those of the terrestrial fauna, the δ15N values were generally higher as it could be expected based on studies on past and modern ecosystems (e.g. Dufour et al. 1999; Drucker et al. 2016). The pond turtle at Noyen-sur-Seine represents another potential source of aquatic food. Although its δ15N values were comparable to the low values found in freshwater fish (7.3 to 8.2 ‰), its δ13C values were drastically lower, ranging from −26.7 to −25.0 ‰. Recent studies have shown that the pond turtle is neither strictly carnivorous nor a pure aquatic animal (Ottonello et al. 2005) and that it favours a woodland environment for its terrestrial movements during nesting (Ficetola et al. 2014).
The otter specimens show a high variability in their δ13C and to a smaller extent in their δ15N values (Table 2, Fig. 2) than was previously observed in Late Mesolithic contexts from the Netherlands (Smits and van der Plicht 2009). This is probably due to a difference in proportion of freshwater and terrestrial resources in their diet. Otters feed not only mainly on fish but consume also amphibians, reptiles (including pond turtles), birds, mammals, and aquatic invertebrates (e.g. Clavero et al. 2003; Lanszki et al. 2006). The species appears thus as a flexible predator that exploited both the terrestrial and freshwater ecosystems as has also been observed in modern ecological studies. Analysis of otter droppings conducted in riverine ecosystems show that fish contributes between ca. 55 and 80 % of the diet (e.g. Lanszki and Molnár 2003; Reid et al. 2013).
δ13C and δ15N values of the human individuals
The human individuals showed δ13C and δ15N values ranging from −21.5 to −20.1 and 11.2 to 12.6 ‰, respectively (Table 3, Fig. 3). The δ13C values collected in two clusters: one around −21.4 ‰ including the individual of Maisons-Alfort, the Noyen-sur-Seine juvenile of layer 9 and the adult of layer 9sup, and another group around −20.2 ‰ including Berry-au-Bac, Cuiry-lès-Chaudardes, Loschbour 1 and the adult of layer 9 of Noyen-sur-Seine. The relatively low 13C cluster showed higher δ15N values (12.1 to 12.6 ‰) than the second one (11.2 to 11.9 ‰), the within-group variability of the δ15N values being less than 0.8 ‰. The two groups can be interpreted as reflecting two different dietary profiles, the lower 13C group having access to aquatic resources since its δ15N values is 6 to 7 ‰ higher in average than the terrestrial animal species. The second cluster possibly incorporated more terrestrial resources leading to lower δ15N values. The large difference in 13C abundances (>3‰) between the human individuals and the pond turtle indicates that this specific animal species was probably not a significant source of food in all of the cases. It is tempting to see a high diet similarity between the individuals of each cluster due to the homogeneity in their isotopic signature. However, the high variability in 13C and 15N abundances displayed by each of the main food groups, namely terrestrial and aquatic, could lead to similar isotopic results despite different diet compositions.
δ13C and δ15N values of bone collagen of the human individuals compared with the terrestrial herbivores represented in mean and standard deviations (red deer, roe deer, wild boar, aurochs), fish and water-dependent animals (turtle, otter), carnivores (wolf, red fox, lynx, wild cat, otter)
δ34S values of preys and the animal predators
The 34S can provide higher resolution of the relative contribution of terrestrial versus freshwater food resources (e.g. Bocherens et al. 2011). However, the number of samples is then reduced since such a measurement requires the combustion of additional collagen material, of which there are not always remainder due to combustion for prior analyses, as well as an excellent preservation of the amino acid containing the element sulphur, namely methionine (Nielsen et al. 1991).
The δ34S values of the terrestrial ungulates plotted between −1.0 and 6.8 ‰, without clear correlation with species (Table 2, Fig. 4). The range of values exhibited by the large ungulates was comparable between the site of Loschbour and Noyen-sur-Seine, which suggests these values are representative of the terrestrial resources for the considered geographical region. This is confirmed by the results obtained on the carnivores, averaging the contribution of prey species, with δ34S values clustering in a similar range as the ungulates of both Noyen-sur-Seine and Loschbour (−1.9 to 9.3 ‰).
δ34S and δ15N values of bone collagen of the red deer (n = 5), roe deer (n = 3), wild boar (n = 3), aurochs (n = 2), fish (n = 5), pond turtle (n = 1), wolf (n = 1), red fox (n = 1), lynx (n = 1), wild cat (n = 1) and otter (n = 2)
In contrast, the species of the aquatic ecosystem displayed quite different δ34S values according to their location. At Noyen-sur-Seine, the pond turtle and the two water-dependent otters provide values ranging from −19.0 to −12.7 ‰, while the Belgian fish clustered between −4.6 and 4.0 ‰ (Table 2, Fig. 4). The δ34S values of the latter were thus overlapping with the terrestrial values provided by the sites of Loschbour 1 and Noyen-sur-Seine. Interestingly, the water-dependent animals of Noyen-sur-Seine were significantly distinct from the terrestrial species of the same site in their lower abundances in 34S. The relatively low δ34S of the otters and pond turtle from Noyen-sur-Seine could reflect the anaeorobic bottom water conditions (reduced sulphur) of the local aquatic ecosystem where they mainly fed, while the higher δ34S values of the Belgian fish may correspond to more oxygenated water conditions (sulphate) (see review in Nehlich 2015). Despite a possible terrestrial contribution to their diet, turtle and otter depend significantly on the riverine environment for food and habitat, and the riparian ecosystem they reflect is very distinct in 34S from the terrestrial forested ecosystem represented by the large hunted ungulates. The Belgian fish shows relatively high 34S abundances, which makes them difficult to distinguish from the terrestrial resources. Two distinct groups of aquatic food are identifiable here and should thus be considered separately as far as the 34S tracking is concerned.
δ34S values of preys and the human individuals
The human δ34S values ranged between −9.9 and 6.6 ‰ (Table 3, Fig. 5). All the humans from Noyen-sur-Seine linked to negative values, ranging from −9.9 to −5.0 ‰, while the other individuals displayed a different range varying from 2.5 to 6.6 ‰. The latter aligned with the terrestrial animal range and partly overlapped with the range of the Belgian fish. The relatively lower δ34S values of the humans from Noyen-sur-Seine placed them in an intermediate position between the terrestrial animals and Belgian fish on the one hand and the riparian animals of the same site on the other. In this last case, the consumption of the local aquatic resources is clearly reflected in the 34S abundances of the human individuals. Outside of Noyen-sur-Seine, the interpretation of the human δ34S values may be hindered by potential insufficient discrimination between the local aquatic and terrestrial resources.
δ34S and δ15N values of bone collagen of the human individuals (n = 7) compared with the terrestrial herbivores including red deer (n = 5), roe deer (n = 3), wild boar (n = 3), aurochs (n = 2) to the terrestrial carnivores including wolf (n = 1), red fox (n = 1), lynx (n = 1), wild cat (n = 1), and the aquatic-dependent animals including Belgian fish (n = 5) and pond turtle (n = 1) and otter (n = 2) from Noyen-sur-Seine
Proportions of consumed prey calculated using SIAR
For each human individual, the possible contribution of each food source was evaluated using SIAR and considered the combined δ13C, δ15N and δ34S values. The food categories were separated between the wild boar, the ruminant ungulates (red deer, roe deer and aurochs), the aquatic-dependent species of the site of Noyen-sur-Seine and the fish from Belgium, since these four groups differ from each other based on at least one of the isotopic systems. The averaged isotopic values of the wild boars and of the ruminant herbivores were calculated from all the available data from Noyen-sur-Seine and Loschbour. The wild boar canine root was excluded from the calculation for δ13C and δ15N values since they may have been affected by a nursing effect, but it was included in the averaged δ34S value due to the lack of trophic effect (e.g. Tanz and Schmidt 2010; Nehlich et al. 2011). Pond turtles were used to calculate the mean δ13C and δ15N values of the riparian resources of the Seine River, and values of the water-dependent otter (NO6500 and NO6600) were also included to estimate the averaged δ34S values, again because of the lack of significant fractionation effect.
The application of the SIAR model allows for the testing of possible combinations of food groups providing dietary protein to animal and human predators over several years. Some difficulties can be foreseen in the overlapping range of δ13C and δ34S values between the Belgian fish and the terrestrial ungulates. It results in a lack of good discrimination between some food groups, namely wild boar and Belgian fish, testified by a significant negative correlation in their posterior distribution for all the tested predators, except for the fox and the otters (Fig. 6). Such a negative correlation was also found between the ruminant herbivores and the wild boar when testing the diet of the wolf and the wild cat. Among the animal predators, the water-dependent otters with relatively low δ13C values displayed a significant consumption of local aquatic resources of Noyen-sur-Seine with ca. 80 % of aquatic-derived protein as the highest likely scenario, while the contribution of the Belgian fish type of resources was very limited in probability. It is consistent with the context-dependent isotopic signature of the freshwater resources and fits the expected maximum contribution of aquatic food observed in ecological studies (e.g. Lanszki and Molnár 2003; Reid et al. 2013). Testing the respective probable contributions of the food resources to the diet of the fox did not provide clear results, especially because none of them were expected to be natural preys of this small predator that is more reliant on small rodents and lagomorphs, unless it has access to the carcasses of larger preys (e.g. Jędrzejewski et al. 1989, 2002; Delibes-Mateos et al. 2007; Helldin and Danielsson 2007). Small preys can be expected to yield lower δ15N values than large ungulates based on the results of previous studies (e.g. Bocherens et al. 2011), and their consumption can thus explain the relatively low δ15N values of the wild cat, which is even more specialized on rodents than the red fox (e.g. Carvalho and Gomez 2001). Interestingly, the wolf is the other animal predator with a low δ15N value, which may also reflect the consumption of ungulates with depleted 15N abundances that can be found among the deer specimens. In the temperate Białowieża forest in Poland, red deer is indeed the most hunted prey of the wolf (e.g. Jędrzejewski et al. 2002). In the same ecosystem, the lynx is specialized on medium-sized preys, such as roe deer (e.g. Okarma et al. 1997). The result of the SIAR simulation for the lynx of Noyen-sur-Seine indicated a diet more oriented towards wild boar, perhaps as a consequence of the high availability of this prey and the interspecific competition with wolf on deer (e.g. Jędrzejewski et al. 1989). In all cases, the contribution of the local riparian resources to the diet of canids and cats of Noyen-sur-Seine was negligible.
Proportional contribution of freshwater fish, riparian resources and terrestrial animals consumed by animal predators as estimated by SIAR using the following assumptions: terrestrial foods were estimated from wild boar and other large herbivores separately, freshwater foods were estimated from Magdalenian and Mesolithic Belgian fish, riparian resources were estimated from pond turtle and otter of Noyen-sur-Seine. Black boxes and whiskers show the median with first and third quartiles and ranges with 1.5 times length of the interquartile range above the third quartile or below the first quartile, respectively. The shaded area indicates the Kernel density plot of the probability density of prey proportions. The brackets link the resources with a significant negative correlation in their posterior distribution
The human individuals of Noyen-sur-Seine were the only ones displaying an unambiguous consumption of local aquatic resources. The local riparian resources were potentially contributing to ca. 30 to 40 % of their dietary protein based on their combined δ13C, δ15N and δ34S values (Figs. 7 and 8). These proportions are in accordance with other estimations based on the amino acid 15N composition (Naito et al. 2013). This implies a significant contribution over a large time span that could fit the hypothesis of Marinval-Vigne et al. (1989) of possible food preservation for a delayed consumption along with fishing activities conducted mainly in summer. On another hand, the contribution of the aquatic resources as based on the Belgian fish appears possible, but with a strong negative correlation with the contribution of wild boar. It means that a higher proportion of one involves a lower proportion of the other. The δ34S values supports the dominant input of the local riparian resources as aquatic-based food, leading to the low probability of the Belgian fish-like resource contribution and then a significant consumption of the wild boar as terrestrial resource.
Proportional contribution of freshwater fish, riparian resources and terrestrial animals consumed by humans as estimated by SIAR using the following assumptions: terrestrial foods were estimated from wild boar and other large herbivores separately, freshwater foods were estimated from Magdalenian and Mesolithic Belgian fish, riparian resources were estimated from pond turtle and otter of Noyen-sur-Seine. Black boxes and whiskers show the median with first and third quartiles and ranges with 1.5 times length of the interquartile range above the third quartile or below the first quartile, respectively. The shaded area indicates the Kernel density plot of the probability density of prey proportions. The brackets link the resources with a significant negative correlation in their posterior distribution
Percentage of prey contribution with the highest probability for the different human individuals. In every case, the wild boar and the Belgian fish are negatively correlated, meaning that a higher proportion of one involves a lower proportion of the other
When aquatic resource consumption is attested, its possible impact on the conventional age has to be considered since the freshwater ecosystem can be a reservoir of older 14C that differs from the 14C of the contemporary atmosphere (e.g. Lanting and van der Plicht 1998). The 14C depletion in freshwater-dissolved inorganic carbon is caused by a restricted exchange in CO2 between water and atmosphere and the input of 14C-depleted sources such as fossil carbonates (Geyh et al. 1998; Lanting and van der Plicht 1998), fossil organic carbon (Boaretto et al. 1998) and ancient glacial melted water (Hall and Henderson 2001). As a result, the offset introduced by the freshwater reservoir effect (FRE) is highly variable and depends on the local causes of the reservoir effect, but does not exceed several hundred years (Keaveney and Reimer 2012; but see exception in Iceland in Ascough et al. 2007). At Noyen-sur-Seine, a difference of 260 ± 63 years was found between the conventional date of one of the aquatic influenced otters (8070 ± 45, NO6500) and a red deer of the same layer (7810 ± 45, NO5000) (Table 1). If the diet accounts for this offset, we can speculate that the consumption of freshwater resources shifted the radiocarbon date of the human remains of no more than ca. 130 years. Indeed, the proportion of the freshwater resources in the human diet was about half of the one estimated in the otter diet. This can thus not explain differences in conventional age as the one observed between the human from layer 9 and the one from layer 9sup at Noyen-sur-Seine.
The consumption of freshwater resources appeared difficult to decipher for the individuals of Berry-au-Bac, Cuiry-lès-Chaudardes, Maison-Alfort and Loschbour 1 using the SIAR model (Fig. 7 and 8). The dietary use of the riparian environment, such as along the Seine bank at Noyen-sur-Seine, can be excluded. However, the non-discriminant results between the Belgian fish and the wild boar as protein providers are a source of uncertainty. In the case of Loschbour 1, the hypothesis of a high consumption of wild boar would fit the archaeological evidences, albeit resulting from old excavations (Cordy 1982) where fish remains may not have been adequately recovered. On the other hand, a necklace made of pike vertebrae found in association with the burial of Cuiry-lès-Chaudardes (Ilett 1998) illustrates the access to aquatic resources, but is certainly not sufficient to provide a clear evidence for fish consumption. If a limited intake of freshwater food could be demonstrated, it would in turn highlight the importance of wild boar in the subsistence pattern of the Loschbour, Cuiry-lès-Chaudardes and Berry-au-Bac individuals, perhaps to a lesser extent for the individual of Maison-Alfort. In this last case, the isotopic discrimination between the different ungulate groups was low, not allowing the determination of the main terrestrial contributor to the human diet. If the dependence on aquatic resources could be established, it would demonstrate that the aquatic environment exploited by humans out of Noyen-sur-Seine was isotopically different from the anoxic riparian context found along the Seine river at that time.
Conclusions
The examined human individuals of the Late Mesolithic from northern France and Luxembourg are all characterized by relatively high δ15N values of their collagen. Due to the complexity of the δ13C and δ15N pattern revealed by the terrestrial as well as the freshwater ecosystem, the interpretation of such values appears difficult at this point. The analysis of the δ34S values in the same collagen allows for a clear distinction between the local riparian and terrestrial resources at Noyen-sur-Seine, and a SIAR model points to a most likely aquatic contribution around 30 to 40 % of the dietary protein. In contrast, the aquatic δ34S values determined on fish specimens of the Mesolithic and older periods of the Magdalenian in Belgium point to a possible isotopic overlap between aquatic and terrestrial ecosystems. As a result, the SIAR model fails to discriminate between this potential aquatic resource and the wild boar contribution for the other studied human individuals. However, the delivered probability reconstruction underlines a difference in protein sources compared with the individuals of Noyen-sur-Seine, despite very comparable results in δ13C and δ15N values that could have been thought to reflect very similar subsistence strategies. This difference could be linked either to the type of aquatic ecosystem exploited for food or to the higher consumption of terrestrial resources, mainly based on wild boar. In contrast, the diet of the human of Maisons-Alfort may have comprised more comparable proportions of the different large ungulates preys.
The analysis of more collagen from local animals could help to constrain the ambiguity introduced by the overlapping δ34S values between aquatic and terrestrial resources. However, the Late Mesolithic burials from Northern France and Luxembourg are generally characterized by a lack of association with animal remains and settlement structures with the notable exception of the extra-funerary remains of the site of Noyen-sur-Seine. The potential of the 34S in bulk collagen as a tracer of aquatic consumption in continental environments should not be neglected, though, but needs a thorough case-by-case preliminary evaluation.
References
Ambrose SH (1990) Preparation and characterization of bone and tooth collagen for isotopic analysis. J Archeol Sci 17(4):431–451
Amundson R, Austin AT, Scuur EAG, Yoo K, Matzek V, Kendall C, Uerbersax A, Brenner D, Baisden WT (2003) Global patterns of the isotopic composition of soil and plant nitrogen. Glob Biogeochem Cycles 17:1031
Arneson LS, MacAvoy SE (2005) Carbon, nitrogen, and sulfur diet-tissue discrimination in mouse tissues. Can J Zool 83(7):989–995
Ascough PL, Cook GT, Church MJ, Dugmore AJ, McGovern TH, Dunbar E, Einarsson Á, Friðriksson A, Gestsdóttir H (2007) Reservoirs and radiocarbon: 14C dating problems in Mývatnssveit, Northern Iceland. Radiocarbon 49:947–961
Auboire G (1991) Les restes humains mésolithiques de Noyen-sur-Seine (Seine-et-Marne, France). l’Anthropologie 95(1):229–236
Ballari SA, Barrios García MN (2014) A review of wild boar Sus scrofa diet and factors affecting food selection in native and introduced ranges. Mammal Rev 44:124–134
Barnes C, Jennings S (2007) Effect of temperature, ration, body size and age on sulphur isotope fractionation in fish. Rapid Commun Mass Spectrom 21:1461–1467
Boaretto E, Thorling L, Sveinbjornsdottir AE, Yechielis Y, Heinemeieri J (1998) Study of the effect of fossil organic carbon on 14C in groundwater from Hvinningdal, Denmark. Radiocarbon 40:915–920
Bocherens H, Drucker D (2003) Trophic level isotopic enrichments for carbon and nitrogen in collagen: case studies from recent and ancient terrestrial ecosystems. Int J Osteoarchaeol 13:46–53
Bocherens H, Billiou D, Patou-Mathis P, Bonjean D, Otte M, Mariotti A (1997) Paleobiological implications of the isotopic signature (13C, 15N) of fossil mammal collagen in Scladina cave (Sclayn, Belgium). Quat Res 48:370–380
Bocherens H, Polet C, Toussaint M (2007) Palaeodiet of Mesolithic and Neolithic populations of Meuse Basin (Belgium): evidence from stable isotopes. J Archaeol Sci 34:10–27
Bocherens H, Drucker DG, Taubald H (2011) Preservation of bone collagen sulphur isotopic compositions in an early Holocene river-bank archaeological site. Palaeogeogr Palaeoclimatol Palaeoecol 310:32–38
Bocherens H, Drucker DG, Madelaine S (2014) Evidence for a 15N positive excursion in terrestrial foodwebs at the Middle to Upper Palaeolithic transition in south-western France: implications for early modern human palaeodiet and palaeoenvironment. J Hum Evol 69:31–43
Bocherens H, Drucker DG, Germonpré M, Lázničková-Galetová M, Naito YI, Wissing C, Brůžek J, Oliva M (2015) Reconstruction of the Gravettian food-web at Předmostí I using multi-isotopic tracking (13C, 15N, 34S) of bone collagen. Quat Int 359:211–228
Bollongino R, Nehlich O, Richards MP, Orschiedt J, Thomas MG, Sell C, Fajkošová Z, Powell A, Burger J (2013) 2000 Years of parallel societies in Stone Age Central Europe. Science 342:479–481
Bonafini M, Pellegrini M, Ditchfield P, Pollard AM (2013) Investigation of the ‘canopy effect’ in the isotope ecology of temperate woodlands. J Archaeol Sci 40:3926–3935
Bonsall C, Cook GT, Hedges REM, Higham TFG, Pickard C, Radovanović I (2004) Radiocarbon and stable isotope evidence of the dietary change from the Mesolithic to the Middle Ages in the Iron Gates: new results from Lepenski Vir. Radiocarbon 46:293–300
Borić D, Grupe G, Peters J, Mikić Ž (2004) Is the Mesolithic-Neolithic subsistence dichotomy real? New stable isotope evidence from the Danube Gorges. Eur J Archaeol 7:221–248
Bosset G, Valentin F (2013) Mesolithic burial practices in the northern half of France: isolated burials and their spatial organization. In: Valentin B, Souffi B, Ducrocq T, Fagnart J-P, Séara F, Verjux C (eds) Palethnographie du Mésolithique : recherches sur les habitats de plein-air entre Loire et Neckar / Mesolithic palethnography : Research on open-air campsites from the river Loire to the Neckar (Actes bilingues de la table-ronde internationale de Paris, 26-27 novembre 2010). Société préhistorique française, Paris, pp 207-216
Bridault A (1997) Broadening and diversification of hunted resources, from the Late Palaeolithic to the late Mesolithic, in the North and East of France and the bordering areas: Old world hunters and gatherers. Anthropozoologica 25–26:295–308
Broadmeadow MSJ, Griffiths H, Maxwell C, Borland AM (1992) The carbon isotope ratio of plant organic material reflects temporal and spatial variations in CO2 within tropical forest formations in Trinidad. Oecologia 89:435–441
Bronk Ramsey C, Lee S (2013) Recent and planned developments of the program OxCal. Radiocarbon 55:720–730
Brooks JR, Flanagan LB, Buchmann N, Ehleringer JR (1997) Carbon isotope composition of boreal plants: functional grouping of life forms. Oecologia 110:301–311
Carvalho JC, Gomez P (2001) Food habits and trophic niche overlap of the red fox, European wild cat and common genet in the Peneda-Gerês National Park. Galemys 13:39–48
Clavero M, Prenda J, Delibes M (2003) Trophic diversity of the otter (Lutra lutra L.) in temperate and Mediterranean freshwater habitats. J Biogeogr 30:761–769
Cordy J-M (1982) La faune mésolithique du gisement de Loschbour près de Reuland (G. D. de Luxembourg). In: Gob A, Spier F (eds) Le Mésolithique entre Rhin et Meuse. Actes du Colloque sur le Paléolithique supérieur final et le Mésolithique dans le Grand-Duché de Luxembourg et dans les régions voisines (Ardenne, Eifel, Lorraine) tenu à Luxembourg le 18 et 19 mai 1981. Publication de la Soc. Préhist. Luxembourgeoise, Luxembourg, pp 119–128
Craine JM, Elmore AJ, Aidar MPM, Bustamante M, Dawson TE, Hobbie EA, Kahmen A, Mack MC, McLauchlan KK, Michelsen A, Nardoto GB, Pardo LH, Peñuelas J, Reich PB, Schuur EAG, Stock WD, Templer PH, Virginia RA, Welker JM, Wright IJ (2009) Global patterns of foliar nitrogen isotopes and their relationships with climate, mycorrhizal fungi, foliar nutrient concentrations, and nitrogen availability. New Phytol 183:980–992
Delibes-Mateos M, Redpath SM, Angulo E, Ferreras P, Villafuerte R (2007) Rabbits as a keystone species in southern Europe. Biol Conserv 137(1):149–156
Delsate D, Guinet JM, Saverwyns S (2009) De l’ocre sur le crâne mésolithique (haplogroupe U5a) de Reuland-Loschbour (Grand-Duché de Luxembourg). Bull Soc Préhist Luxembourgeoise 31:7–30
Delsate D, Brou L, Spier F (2011) L’inhumation mésolithique de Loschbour (Loschbour 1) - résultats des analyses récentes. In: Sous nos pieds/Unter unseren Füßen: Archäologie in Luxemburg - Archéologie au Luxembourg - 1995-2010. Musée National d’Histoire et d’Art, Luxembourg, pp 138-142
DeNiro MJ (1985) Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature 317:806–809
Drucker D (2001) Validation méthodologique de l’analyse isotopique d’ossements fossiles et apports aux reconstitutions paléoécologiques du Paléolithique supérieur du sud-ouest de la France. Dissertation, University of Paris 6 Pierre et Marie Curie
Drucker DG, Bocherens H, Billiou D (2003) Evidence for shifting environmental conditions in Southwestern France from 33,000 to 15,000 years ago derived from carbon-13 and nitrogen-15 natural abundances in collagen of large herbivores. Earth Planet Sci Lett 216:163–173
Drucker DG, Bridault A, Hobson KA, Szuma E, Bocherens H (2008) Can collagen carbon-13 abundance of large herbivores reflect the canopy effect in temperate and boreal ecosystems? Evidence from modern and ancient ungulates. Palaeogeogr Palaeoclimatol Palaeoecol 266:69–82
Drucker DG, Bridault A, Cupillard C, Hujic A, Bocherens H (2011) Evolution of habitat and environment of red deer (Cervus elaphus) during the Late-glacial and early Holocene in eastern France (French Jura and the western Alps) using multi-isotope analysis (δ13C, δ15N, δ18O, δ34S) of archaeological remains. Quat Int 245:268–278
Drucker DG, Hobson KA, Münzel SC, Pike-Tay A (2012) Intra-individual variation in stable carbon (δ13C) and nitrogen (δ15N) isotopes in mandibles of modern caribou of Qamanirjuaq (Rangifer tarandus groenlandicus) and Banks Island (Rangifer tarandus pearyi): implications for tracing seasonal and temporal changes in diet. Int J Osteoarchaeol 22:494–504
Drucker DG, Rosendahl W, Van Neer W, Weber M-J, Görner I, Bocherens H (2016) Environment and subsistence in northwestern Europe during the Younger Dryas: an isotopic study of the human of Rhünda (Germany). J Archaeol Sci Rep 6:690–699
Ducrocq T, Ketterer I (1995) Le gisement mésolithique du “Petit Marais”, La Chaussée-Tirancourt (Somme). Bull Soc préhist fr 92:249–259
Ducrocq T, Bridault A, Coutard S (2008) Le gisement mésolithique de Warluis: approche préliminaire. Mém Soc préhist fr 45:85–106
Dufour E, Bocherens H, Mariotti A (1999) Palaeodietary implications of isotopic variability in Eurasian lacustrine fish. J Archaeol Sci 26:627–637
Ficetola GF, Padoa-Schioppa E, Monti A, Massa R, De Bernardi F, Bottoni L (2014) The importance of aquatic and terrestrial habitat for the European pond turtle (Emys orbicularis): implications for conservation planning and management. Can J Zool 82:1704–1712
Fischer A, Olsen J, Richards M, Heinemeier J, Sveinbjörnsdóttir ÁE, Bennike P (2007) Coast-inland mobility and diet in the Danish Mesolithic and Neolithic: evidence from stable isotope values of humans and dogs. J Archaeol Sci 34:2125–2150
Francey RJ, Gifford RM, Sharkey TD, Weir B (1985) Physiological influences on carbon isotope discrimination in huon pine (Lagarostrobos franklinii). Oecologia 66:211–218
Fry B, Gest H, Hayes JM (1986) Sulfur isotope effects associated with protonation of HS- and volatilization of H2S. Chem Geol (Isot Geosci Sect) 58(3):253–258
Geyh MA, Schotterer U, Grosjean M (1998) Temporal changes of the 14C reservoir effect in lakes. Radiocarbon 40:921–931
Gob A (1982) L’occupation mésolithique de l’abri du Loschbour près de Reuland (G.-D. de Luxembourg). In: Gob A, Spier F (eds) Le Mésolithique entre Rhin et Meuse, Actes du colloque sur le Paléolithique final et le Mésolithique dans le Grand-Duché de Luxembourg et dans les régions voisines (Ardenne, Eifel, Lorraine), Luxembourg, 18 et 19 mai 1981. Société préhistorique luxembourgeoise, Luxembourg, pp 91–117
Guiry EJ, Hillier M, Richards MP (2015) Mesolithic Dietary Heterogeneity on the European Atlantic Coastline. Curr Anthropol 56(3):460–70
Hall BL, Henderson GM (2001) Use of uranium-thorium dating to determine past 14C reservoir effects in lakes: examples from Antarctica. Earth Planet Sci Lett 193:565–577
Helldin JO, Danielsson AV (2007) Changes in red fox Vulpes vulpes diet due to colonisation by lynx Lynx lynx. Wildl Biol 13(4):475–480
Higham TFG, Bronk Ramsey C, Brock F, Baker D, Ditchfield P (2007) Radiocarbon dates from the Oxford AMS system archaeometry datelist 32. Archaeometry 49:1–60
Holmer M, Storkholm P (2001) Sulphate reduction and sulphur cycling in lake sediments: a review. Freshw Biol 46(4):431–451
Ilett M (1998) Cuiry-lès-Chaudardes, les Fontinettes. Bilan scientifique de la région Picardie. Service régional de l’Archéologie, Amiens, pp 26–27
Jędrzejewski W, Jędrzejewska B (1992) Foraging and diet of the red fox Vulpes vulpes in relation to variable food resources in Biatowieza National Park, Poland. Ecography 15(2):212–220
Jędrzejewski W, Jędrzejewska B, Szymura A (1989) Food niche overlaps in a winter community of predators in Białowieża primeval forest, Poland. Acta Theriol 34:487–496
Jędrzejewski W, Schmidt K, Theuerkauf J, Jędrzejewska B, Selva N, Zub K, Szymura L (2002) Kill rates and predation by wolves on ungulate populations in Białowieża primeval forest (Poland). Ecology 83:1341–1356
Katzenberg MA, Bazaliiskii VI, Goriunova OI, Savel’ev NA, Weber AW (2010) Diet reconstruction of prehistoric hunter-gatherers in the Lake Baikal Region. Prehistoric hunter-gatherers of the Baikal Region, Siberia. University of Pennsylvania Museum of Archaeology and Anthropology, Philadelphia, pp 175–191
Keaveney EM, Reimer PJ (2012) Understanding the variability in freshwater radiocarbon reservoir offsets: a cautionary tale. J Archaeol Sci 39:1306–1316
Lanszki J (2005) Diet composition of red fox during rearing in a moor: a case study. Folia Zool 54(1/2):213–216
Lanszki J, Molnár T (2003) Diet of otters living in three different habitats in Hungary. Folia Zool 52:378–388
Lanszki J, Molnár M, Molnár T (2006) Factors affecting the predation of otter (Lutra lutra) on European pond turtle (Emys orbicularis). J Zool 270:219–226
Lanting JN, van der Plicht J (1998) Reservoir effects and apparent 14C-ages. J Irish Archaeol 9:151–165
Leduc C, Bridault A, Souffi B, David E, Drucker DG (2013) Apports et limites de l’étude des vestiges fauniques à la caractérisation d’un site mésolithique de plein air à Paris : « 62 rue Henry-Farman » (15e arrondissement). Bull Soc préhist fr 110:257–280
Léotard JM, Straus LG, Otte M (1999) L’Abri du Pape. Bivouacs, enterrements et cachettes sur la Haute Meuse belge: du Mésolithique au Bas Empire Romain. Etudes et Recherches Archéologiques de l’Université de Liège, ERAUL
Lightfoot E, Boneva B, Miracle PT, Šlaus M, O’Connell TC (2011) Exploring the Mesolithic and Neolithic transition in Croatia through isotopic investigations. Antiquity 85:73–86
Lillie MC, Richards M (2000) Stable isotope analysis and dental evidence of diet at the Mesolithic-Neolithic transition in Ukraine. J Archaeol Sci 27:965–972
Lillie M, Budd C, Potekhina I (2011) Stable isotope analysis of prehistoric populations from the cemeteries of the Middle and Lower Dnieper Basin, Ukraine. J Archaeol Sci 38(1):57–68
Longin R (1971) New method of collagen extraction for radiocarbon dating. Nature 230:241–242
Lubell D, Jackes M, Schwarcz H, Knyf M, Meiklejohn C (1994) The Mesolithic–Neolithic transition in Portugal: isotopic and dental evidence of diet. J Archaeol Sci 21:201–216
Marinval-Vigne M-C, Mordant D, Auboire G, Augereau A, Bailon S, Dauphin C, Delibrias G, Krier V, Leclerc A-S, Leroyer C, Marinval P, Mordant C, Rodriguez P, Vilette P, Vigne J-D (1989) Noyen-sur-Seine, site stratifié en milieu fluviatile: une étude multidisciplinaire intégrée. Bull Soc Préhi Fr 86:370–379
McCutchan JH, Lewis WM, Kendall C, McGrath CC (2003) Variation in trophic shift for stable isotope ratios of carbon, nitrogen, and sulfur. Oikos 102(2):378–90
Meiklejohn C, Bosset G, Valentin F (2010) Radiocarbon dating of Mesolithic human remains in France. Mesolithic Miscellany 21:10–56
Meiklejohn C, Miller R, Toussaint M (2014) Radiocarbon dating of Mesolithic human remains in Belgium and Luxembourg. Mesolithic Miscellany 22:10–39
Mordant D, Mordant C (1992) Noyen-sur-Seine: a Mesolithic waterside settlement. In: Coles B (ed) The Wetland Revolution in Prehistory. The Prehistoric Society, Exeter, pp 55–64
Mordant D, Valentin B, Vigne J-D (2013) Noyen-sur-Seine, twenty-five years on. In: Valentin B, Souffi B, Ducrocq T, Fagnart J-P, Séara F, Verjux C (eds) Palethnographie du Mésolithique : recherches sur les habitats de plein-air entre Loire et Neckar / Mesolithic palethnography : Research on open-air campsites from the river Loire to the Neckar (Actes bilingues de la table-ronde internationale de Paris, 26-27 novembre 2010). Société préhistorique française, Paris, pp 37-49
Naito YI, Chikaraishi Y, Ohkouchi N, Drucker DG, Bocherens H (2013) Nitrogen isotopic composition of collagen amino acids as an indicator of aquatic resource consumption: insights from Mesolithic and Epipalaeolithic archaeological sites in France. World Archaeol 45:338–359
Nehlich O (2015) The application of sulphur isotope analyses in archaeological research: a review. Earth-Sci Rev 142:1–17
Nehlich O, Richards MP (2009) Establishing collagen quality criteria for sulphur isotope analysis of archaeological bone collagen. Archaeol Anthropol Sci 1:59–75
Nehlich O, Borić D, Stefanović S, Richards MP (2010) Sulphur isotope evidence for freshwater fish consumption: a case study from the Danube Gorges, SE Europe. J Archaeol Sci 37:1131–1139
Nehlich O, Fuller BT, Jay M, Mora A, Nicholson RA, Smith CI, Richards MP (2011) Application of sulphur isotope ratios to examine weaning patterns and freshwater fish consumption in Roman Oxfordshire, UK. Geochim Cosmochim Acta 75(17):4963–4977
Nielsen H, Pilot J, Grinenko LN, Grinenko VA, Lein AY, Smith JW, Pankina RG (1991) Lithospheric sources of sulfur. In: Krouse HR, Grinenko VA (eds) Stable Isotopes in the Assessment of Natural and Anthropogenic Sulfur in the Environment. John Wiley & Sons, Chichester, pp 65–132
Okarma H, Jędrzejewski W, Schmidt K, Kowalczyk R, Jędrzejewska B (1997) Predation of Eurasian lynx on roe deer in Białowieża primeval forest, Poland. Acta Theriol 42:203–224
Ottonello D, Salvidio S, Rosecchi E (2005) Feeding habits of the European pond terrapin Emys orbicularis in Camargue (Rhône delta, Southern France). Amphibia-Reptilia 26:562–565
Parnell AC, Inger R, Bearhop S, Jackson AL (2010) Source partitioning using stable isotopes: coping with too much variation. PLoS ONE 5, e9672
Peterson BJ, Fry B (1987) Stable isotopes in ecosystem studies. A Rev Ecol Syst 18:293–320
Posth C, Renaud G, Mittnik A, Drucker DG, Rougier H, Cupillard C, Valentin F, Thevenet C, Furtwängler A, Wißing C, Francken M, Malina M, Bolus M, Lari M, Gigli E, Capecchi G, Crevecoeur I, Beauval C, Flas D, Germonpré M, van der Plicht J, Cottiaux R, Gély B, Ronchitelli A, Wehrberger K, Grigorescu D, Svoboda J, Semal P, Caramelli D, Bocherens H, Harvati K, Conard NJ, Haak W, Powell A, Krause J (2016) Pleistocene mitochondrial genomes suggest a single major dispersal of non-Africans and a late glacial population turnover in Europe. Curr Biol 26:1–7
Poupin N, Bos C, Mariotti F, Huneau J-F, Tomé D, Fouillet H (2011) The nature of the dietary protein impacts the tissue-to-diet 15N discrimination factors in laboratory rats. PLoS ONE 6, e28046
Privat KL, O’Connell TC, Hedges RE (2007) The distinction between freshwater-and terrestrial-based diets: methodological concerns and archaeological applications of sulphur stable isotope analysis. J Archaeol Sci 34(8):1197–1204
Reid N, Thompson D, Hayden B, Marnell F, Montgomery WI (2013) Review and quantitative meta-analysis of diet suggests the Eurasian otter (Lutra lutra) is likely to be a poor bioindicator. Ecoll Indicators 26:5–13
Reimer PJ, Bard E, Bayliss A, Beck JW, Blackwell PG, Bronk Ramsey C, Grootes PM, Guilderson TP, Haflidason H, Hajdas I, Hatte C, Heaton TJ, Hoffmann DL, Hogg AG, Hughen KA, Kaiser KF, Kromer B, Manning SW, Niu M, Reimer RW, Richards DA, Scott EM, Southon JR, Staff RA, Turney CSM, Van der Plicht J (2013) IntCal13 and Marine13 radiocarbon age calibration curves 0-50,000 years cal BP. Radiocarbon 55:1869–1887
Richards MP, Mellars PA (1998) Stable isotopes and the seasonality of the Oronsay middens. Antiquity 72(275):178–184
Richards MP, Fuller BT, Hedges REM (2001) Sulphur isotopic variation in ancient bone collagen from Europe: implications for human palaeodiet, residence mobility, and modern pollutant studies. Earth Planet Sci Lett 191:185–190
Richards MP, Douglas TD, Koch E (2003a) Mesolithic and Neolithic subsistence in Denmark: new stable isotope data. Curr Anthropol 44:288–295
Richards MP, Fuller BT, Sponheimer M, Robinson T, Ayliffe L (2003b) Sulphur isotopes in palaeodietary studies: a review and results from a controlled feeding experiment. Int J Osteoarchaeol 13(1-2):37–45
Richards MP, Schulting RJ, Hedges RE (2003c) Archaeology: sharp shift in diet at onset of Neolithic. Nature 425(6956):366–366
Robbins CT, Felicetti LA, Florin ST (2010) The impact of protein quality on stable nitrogen isotope ratio discrimination and assimilated diet estimation. Oecologia 162:571–579
Schoeninger MJ, DeNiro MJ (1984) Nitrogen and carbon isotopic composition of bone collagen from marine and terrestrial animals. Geochim Cosmochim Acta 48:625–639
Schulting RJ, Richards MP (2001) Dating women and becoming farmers: new palaeodietary and AMS dating evidence from the Breton Mesolithic cemeteries of Téviec and Hoëdic. J Anthropol Archaeol 20(3):314–344
Smits L, van der Plicht H (2009) Mesolithic and Neolithic human remains in the Netherlands: physical anthropological and stable isotope investigations. J Archaeol Low Countries 1:55–85
Tanz N, Schmidt H (2010) δ34S value measurements in food origin assignments and sulfur isotope fractionations in plants and animals. J Agric Food Chem 58:3139–3146
Team R Core (2013) R: a language and environment for statistical computing. 409
Toussaint M, Brou L, Le Brun-Ricalens F, Spier F (2009) The Mesolithic site of Heffingen-Loschbour (Grand Duchy of Luxembourg). A yet undescribed human cremation possibly from the Rhine-Meuse-Schelde culture: anthropological, radiometric and archaeological implications. In: Crombé P, Van Strydonck M, Sergant J, Boudin M, Bats M (eds) Chronology and evolution within the Mesolithic of North-West Europe. Cambridge Scholars Publishing, Cambridge, pp 239–260
Valentin F, Cottiaux R, Buquet-Marcon C, Confalonieri J, Delattre V, Lang L, Le Goff I, Lawrence-Dubovac P, Verjux C (2008) Découvertes récentes d‘inhumations et d’incinération datées du Mésolithique en Ile de France. Revue Archéologique d’Ile-de-France 1:21–42
van der Merwe NJ, Medina E (1991) The canopy effect, carbon isotope ratios and foodwebs in Amazonia. J Archaeol Sci 18:249–259
Acknowledgements
We acknowledge the financial support provided by the PCR ‘Paléolithique final et Mésolithique dans le Bassin Parisien et ses marges’ (dir. B. Valentin). The European Social Fund and the Ministry of Science, Research and Arts of Baden-Württemberg funded the current position of D.G. Drucker. The contribution of Wim Van Neer to this paper presents research results of the Interuniversity Attraction Poles Programme—Belgian Science Policy. Thanks are due to A. Bridault, M.-C. Marinval-Vigne, H. Bocherens and D. Billiou for the initial sampling of the animal remains of Noyen-sur-Seine. We are grateful to Michaël Ilett, Lamys Hachem and Bruno Robert (Trajectoires, CNRS-UMR8215) for allowing the study of Cuiry-les-Chaudardes and Berry-au-Bac. We thank Foni Le Brun-Ricalens, Laurent Brou, François Valotteau (Centre National de Recherche Archéologique, Luxembourg), Jean-Michel Guinet, Edmée Engel and Alain Faber (Musée National d’Histoire Naturelle, Luxembourg) for allowing sampling of faunal remains from the Loschbour 1 excavation. The isotopic analysis benefited from the technical support of Bernd Steinhilber, Catherine Bauer, Christoph Wißing and the team of Biogeologie (Department of Geoscience, University of Tübingen). We are grateful to Thomas Tütken and an anonymous reviewer for their valuable comments. Furthermore, we wish to thank Sophia Haller for English proofreading.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Drucker, D.G., Valentin, F., Thevenet, C. et al. Aquatic resources in human diet in the Late Mesolithic in Northern France and Luxembourg: insights from carbon, nitrogen and sulphur isotope ratios. Archaeol Anthropol Sci 10, 351–368 (2018). https://doi.org/10.1007/s12520-016-0356-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12520-016-0356-6