Trzciniec cultural circle. An overview

Studies based on site catchment analyses have allowed the settlement model of TCC communities in Kuyavia to be reconstructed [26]. It was organized around so-called micro-regions comprising central settlements, smaller settlements (campsites) and a neighbouring cemetery. Most of the human activity took place within a radius of ca. 2 km from the central settlements. Most plausibly, after their emergence in the northern lowlands ca. 1800 BC, “Trzciniec” people migrated south-east, colonised the uplands of today’s south-eastern Poland and western Ukraine, keeping those settlement patterns developed in the north. The mainstays of their subsistence strategy, both in lowlands and uplands, was animal husbandry and plant cultivation. This is confirmed by grazing markers visible in the pollen records of various landscape zones: on wet floodplain meadows of river terraces, on dry meadows in the deforested parts of areas of high relief, on the edges of forests and in forest clearings. Ruminants such as cattle and sheep/goat dominated the structure of the herds while pigs and other domesticates were less important [22, 26]. The presence of cereal pollen in palynological profiles indicates a complementary role of crops, mainly wheat and barley [22, 43]. Direct radiocarbon dating of charred broomcorn millet grains and stable isotope analyses of human remains revealed that since the middle of 15^th c. BC this plant was cultivated and consumed in the uplands while in the lowlands it was most probably unknown before the 13^th c. BC [44, 45].

The characteristic form of burial rite among TCC societies was collective graves [22, 46–48]. The deceased were placed in antithetic (antipodal) order, with heads on the shorter side of a pit and legs pointing to its centre (Fig 2). Human bones in such graves are usually found arranged in piles, and only some skeletons or parts of skeletons were kept in anatomical order. How such arrangements were built is difficult to reconstruct because of either lack of accurate field documentation or the inability to distinguish between human agency and post-depositional processes. It seems, however, that bodies were placed in the graves shortly after death, i.e., with preserved soft tissues and joint connections. This was preceded by moving the remains of previously buried individuals, likely at various stages of decomposition, to make space for the new interments. Radiocarbon dating of tombs from numerous cemeteries in Poland and western Ukraine indicates that the collective graves were in use from periods of 10 years to a maximum of 200–250 years [47, 49, 50]. They could therefore be used by several to even a dozen generations or more.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Simplified outlines of graves nos.

69 (A) and 99 (B) in Żerniki Górne (after [51]) and nos. 4 (C) and 5 (D) in Gustorzyn (after [48], simplified). Human and animal bones are marked in yellow. In Gustorzyn darker shade marks bones described by excavators as belonging to human skulls, in Żerniki Górne human skulls were solely recognised basing on published figures; stone constructions were omitted.

https://doi.org/10.1371/journal.pone.0282472.g002

Gustorzyn

The cemetery at Gustorzyn site 1 is located within the Kuyavia Plain, known also as the Kuyavia Moraine Plateau. It lies at the border of the villages of Gustorzyn and Polówka, on a promontory of a flat moraine plateau, separated from the south by the Zgłowiączka river valley and from the west and north by a small valley with a tributary to the Zgłowiączka. Two communal graves (nos. 4 and 5; Fig 2) were discovered during rescue excavations in 1981 on the north-western outskirts of an active gravel pit [52]. In total, at least 32 people were buried here but it is obvious that the original number of individuals was higher. The general characteristic of the two preserved graves is presented in S1 Text.

Eight of the 17 individuals included in this study were ¹⁴C-dated by Accelerator Mass Spectrometry (AMS) and the remaining collagen was submitted for carbon (δ¹³C) and nitrogen (δ¹⁵N) stable isotope analyses. The results have been published elsewhere [45, 53] and summarised in Table 1. They indicate that both graves were in use between approx. middle 15^th and early 13^th centuries BC, and that the diet of dated individuals was based mainly on C₃ plants and terrestrial animals. For both radiocarbon dating and stable isotope analyses, outer layers of petrosal bones were sampled. Inner parts would normally reflect diet in the early years of life [53]. However, the outer parts most probably inform about later stages of infancy, most certainly after weaning [54], as evidenced by relatively low δ¹⁵N values (≤ 10.8 ‰) (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Detailed context information, strontium (⁸⁷Sr/⁸⁶Sr) and oxygen (δ¹⁸O_P-VSMOW) isotope data, AMS radiocarbon ages and bone bulk collagen carbon (δ¹³C) and nitrogen (δ¹⁵N) isotopic data for humans from cemeteries in Gustorzyn and Żerniki Górne.

Anthropological data for Gustorzyn after [48, 56]. All ¹⁴C dates were calibrated with OxCal 4.4 [123] using the IntCal 2020 curve [124].

https://doi.org/10.1371/journal.pone.0282472.t001

Żerniki Górne

The cemetery in Żerniki Górne site 1 is located in the central Nida Basin, which is a part of the Lesser Poland Upland. The site is located on a promontory of a loess plateau, falling south towards a small valley, drained by a little watercourse and surrounded by ridges of limestone formations (Fig 5). It was excavated from 1965 to 1968 [51]. The TCC necropolis was established on the location of older cemeteries, used firstly by Final Neolithic Corded Ware culture communities and then by Early Bronze Age Mierzanowice culture groups [51, 55]. At the top and partly on the slopes of the hill, four stone circles (mainly blocks of marl deposits) with a diameter of ca. 4.0 to 6.0 m have been laid. Near them, five pits filled with burnt animal bones and fragments of ceramic vessels were unearthed. The stone circles and pits were covered by a mound (the assumed central burial was destroyed by a modern pit), with the oldest graves located nearby and younger ones dug into it [47]. In total, 16 graves were discovered in which at least 170 individuals were buried. The five graves chosen for this study are described in more detail in S1 and graves nos. 69 and 99 are shown on Fig 2. The Bayesian modelling of 25 ¹⁴C AMS dates indicates that the cemetery was used for 140 to 310 years, i.e., for a maximum of eight to ten generations [45].

Between one to three individuals from each grave were AMS ¹⁴C-dated and targeted with δ¹³C and δ¹⁵N analyses (Table 1) [45, 47]. In all but one case, the outer parts of petrosal bones were sampled. The radiocarbon-dated humans lived between the middle 16^th and late 15^th centuries BC, except one from the end of the 14^th century BC. As in Gustorzyn, their main source of dietary protein was C₃ plants and animals with no evidence of admixture of millet and no signal of breastfeeding was detected among the children studied [45]. However, it should be noted that ten more individuals from Żerniki Górne were AMS ¹⁴C-dated and one of them (individual 1 from grave no. 54), who died in the second half of the 14^th century BC, was a “millet-eater” [45].

Anthropological analyses

Human remains from graves nos. 4 and 5 in Gustorzyn were first anthropologically analysed in the early 2000’s [56] while finds from Żerniki Górne were investigated by A. and A. Wierciński and published in an archaeological monography of the site [51]. These analyses were later verified during material sampling for this study. The anthropological examination was prepared according to the classical morphological methods with the special modification for collective graves. The first stage was a selection of bones belonging to one individual. As a result of the similar age at death for some individuals, it was not possible to assign all remains. After the separation of individual skeletons, the anthropological determination was conducted as for single burials. In the case of child skeletons, the analysis also included the degree of ossification and the size of long bones [57, 58], as well as the development of dentition [59]. The age of adult individuals was established on the basis of the fusing of cranial sutures and dental attrition. Sex was determined only for adult individuals and was based on a comprehensive analysis of dimorphic features on the skull and pelvis [60].

Strontium isotope analyses

The principles of using Sr isotopes as a tool for provenancing archaeological individuals have been described elsewhere [61–64] and thus only a summary is presented here. In provenance studies, the ⁸⁷Sr/⁸⁶Sr ratio is of primary interest as it does not change (fractionate) in any measurable way when Sr is transferred from the local source rocks through the natural environment and biosphere [65, 66]. However, because various components of the environment usually differ from each other in their ⁸⁷Sr/⁸⁶Sr ratios [67, 68], the Sr isotope composition of organic tissues (plants, animals, and humans) results from the mixing of strontium derived from local Sr reservoirs, i.e., the geological substrate and surface water (and/or groundwater). The Sr isotope composition of human enamel, for example, represents a weighted average of Sr derived from all the food and water ingested while the enamel was mineralized [69, 70], i.e., in the case of molars, in childhood. Thus, human enamel reaches an ⁸⁷Sr/⁸⁶Sr ratio, which is between the composition of the geological substrate and the drinking water [61, 64, 71] and is usually very close to that of the water.

To identify local vs. non-local individuals in a studied human population it is necessary to establish the range of “local” Sr isotope ratios available to plants, animals and humans within a given ecosystem. This can be done by analysing the Sr isotope composition of the geological substrate (rocks, sediments, soils), surface and/or ground waters, and samples of plants and animals. Recent studies, however, showed that the Sr isotope system of modern environments, i.e., its surface waters, flora and fauna, can be influenced by inputs of Sr from various anthropogenic sources [67, 72–74]. Therefore, caution is needed if Sr isotope signatures of modern samples are used. Alternatively, subfossil samples of animals (“archaeofauna”) can be analysed. This approach is usually hampered by the availability of animal remains from settlements located close to cemeteries. Moreover, domestic archaeological animals may have been imported to the studied sites and therefore a non-local provenance of “archaeofauna” can never be excluded beforehand [61, 71, 75].

In the case of the two sites covered by this study samples of archaeological fauna were not available to us. Hence, we estimated the local Sr isotope background by (i) reviewing the geological substrates of both cemeteries and their surrounding areas, (ii) reviewing published data on Sr isotope composition of sediments and waters in both regions, (iii) reviewing published data on Sr isotope composition of archaeological humans and animals from other sites in both regions.

Strontium isotope analyses were carried out in the Isotope Laboratory at the Adam Mickiewicz University in Poznań, Poland. The mechanically isolated enamel was first cleaned in an ultrasonic bath in ultrapure water to remove sediment particles. Afterwards, about 10 mg of powdered enamel was treated sequentially with 2 ml of 0.1 N ultrapure acetic acid (5 times) to eliminate the diagenetic Sr contamination, according to the procedure described by [76]. Subsequently, the samples were dissolved on a hot plate (~100°C, overnight) in closed PFA vials using 1 N HNO₃. The miniaturized chromatographic technique developed by [77] and modified by [78] was applied for Sr separation. Purified Sr was loaded with a TaCl₅ activator on a single Re filament and analyzed in dynamic multi-collection mode on a Finnigan MAT 261 mass spectrometer. Samples were corrected for mass-dependent fractionation by internal normalisation to a fixed ⁸⁶Sr/⁸⁸Sr of 0.1194 using an exponential mass bias law [79, 80]. During this study, the NIST SRM 987 Sr standard yielded ⁸⁷Sr/⁸⁶Sr = 0.710224 ± 11 (2σ, n = 15). The measured ratios were normalized to a nominal value of 0.710240 for the standard NIST SRM 987. Total procedure blanks were less than 70 pg.

Oxygen isotope analyses

The principles of using O isotopes in archaeological provenance studies have been detailed elsewhere [81, 82] and only a summary is given here. The O isotope composition (expressed as δ¹⁸O) in animal and human tissues is related to the isotopic composition of ingested water, primarily controlled by the local meteoric water composition, with minor contributions from food, usually at the level of 30% [83–88].

δ¹⁸O in local meteoric water are products of fractionation in the hydrological cycle and are dependent mainly on local temperature and, to a lesser degree, on local altitude, latitude, humidity and distance from the sea [89–97]. Hence, if the measured δ¹⁸O values of human tissues are different than the δ¹⁸O values of local meteoric water, the non-local origins of this human can be assessed. However, as δ¹⁸O values in precipitation are correlated with temperature they can vary seasonally within the same geographic area [91]. Moreover, they can be affected by long-term climate changes. As a consequence, caution is needed when comparing oxygen isotope ratios between humans or animals living in the same location but in different geological and archaeological periods. Consequently, δ¹⁸O values measured for mineralized human and animal tissues can serve for reconstructing past climatic conditions (palaeotemperatures) [98–101].

A common approach is to compare δ¹⁸O values measured for archaeological animals and humans to modern δ¹⁸O precipitation values, as δ¹⁸O precipitation values reconstructed for past climates are rare. It is, however, advised to use such values only as approximation as they may not reflect the past values accurately [81].

In provenance studies using dental enamel and bone one measures the relative abundance of ¹⁶O and ¹⁸O in the sample (¹⁸O/¹⁶O) and this is typically expressed in permil notation (‰) relative to standard composition using the following formula: (1)

The bioapatite lattice in bones and tooth enamel of humans and other mammals contains two main locations where oxygen atoms are found. These are the oxygen bound to the phosphate group (PO₄^3–) and that bound to the structural carbonate (CO₃^2–), with an additional minor component of hydroxyl (OH^–) oxygen. Both the phosphate and carbonated bound oxygen can be isolated and used for isotope analysis and are commonly reported as δ¹⁸O_P and δ¹⁸O_C, respectively [99, 102–105]. δ¹⁸O_P is sourced from body water, δ¹⁸O_C comes from blood biocarbonate (δ¹⁸O_BC) which is isotopically equilibrated with body water and depends on the temperature at which it is precipitated [86, 87, 95, 106–108].

The human body isotopically fractionates oxygen between ingested water and body water [83, 86, 109]. However, the fractionation appears to be linear, and there is a consistent offset of ~18 ‰ between the δ¹⁸O of body water and PO₄ in bioapatite, and a further offset of ~8 ‰ between PO₄ and CO₃ bound oxygen [99, 106]. Nevertheless, deviations have been reported and they have been attributed to, for instance, differences in food sources or water sources and volume [83, 85, 86, 95, 108, 110] or processing the drinking water by stewing, boiling, fermentation or distillation [111]. Most importantly, isotopic fractionation between the mother’s body and breast milk leads to elevated O isotope values in tissues forming during childhood, and the reported ¹⁸O enrichment ranges from 0.5 to ~2 ‰ [112, 113].

Oxygen isotope nomenclature can be further complicated by the two different isotopic scales that have been used for carbonates and waters. Historically, δ¹⁸O values were first expressed relative to PDB (Pee Dee Belemnite) carbonate standard which is now exhausted. Today two other standards are used, VPDB (Vienna Pee Dee Belemnite) and more common VSMOW (Vienna Standard Mean Ocean Water). The relationship between the two isotope scales is well understood and most authors use Coplen’s [114] conversion equation: (2)

The relationship between δ¹⁸O_P values in human bioapatite and δ¹⁸O values in drinking water (referred to as δ¹⁸O_DW) is well established [87, 96, 115]. Daux and collaborators [109] provide a series of equations for calculating drinking water oxygen isotope values from measured δ¹⁸O_P values using either (i) tap water values or (ii) OIPC (Online Isotopes in Precipitation Calculator) [115] precipitation values. Combining this data with previously published work [86, 95, 116] provides a single equation for converting between δ¹⁸O_P and δ¹⁸O_DW: (3)

Pollard and collaborators in their meta-study [117] provide another equation termed “Classical regression” which gives a regression with a very similar slope and intercept: (4)

A direct relationship between δ¹⁸O_P and δ¹⁸O_C from human tooth enamel has been established by a study of Chenery and collaborators [81]. They provide a conversion equation from δ¹⁸O_C to δ¹⁸O_P: (5)

Combining (3) and (5) allows for the direct conversion from enamel δ¹⁸O_C to δ¹⁸O_DW: (6)

In this paper, all oxygen isotope data are reported on the VSMOW scale. For literature data, oxygen isotope data reported on the VPDB scale are first converted to VSMOW scale using (2) and δ¹⁸O_C values are converted to δ¹⁸O_P values using (5). Following this, δ¹⁸O_P values are converted to δ¹⁸O_DW values using (3).

However, whilst the conversion equations between bioapatite δ¹⁸O values and drinking water δ¹⁸O values described above can be easily implemented, the full treatment of the data requires that the uncertainty (error) in the regression lines be propagated into the final uncertainty on the δ¹⁸O_DW. When this is considered, it is clear that final calculated δ¹⁸O_DW values have a relatively large error, above 1 ‰, in some cases reaching ± 3.5 ‰ [109, 117]. Such errors are substantially higher than the analytical precision that can be obtained on replicated δ¹⁸O of bioapatite samples (0.03 to 0.2 ‰ 1σ) and in some cases may well be higher than δ¹⁸O variability within local human populations or even mask outstanding values received for non-local individuals. Hence, as suggested in the literature [117–119] to study human mobility within and between populations only δ¹⁸O values measured directly in bioapatite are compared and predicted δ¹⁸O values in precipitation are used only as a background reference.

In this study δ¹⁸O_P analyses were conducted in the GeoZentrum Nordbayern at the Friedrich-Alexander University in Erlangen-Nürnberg, Germany. Powdered enamel remaining after analyses in Poznań was first transferred to Erlangen. Chemical conversion of apatite bound phosphate into trisilverphosphate (Ag₃PO₄) was performed according to the method described by Joachimski and collaborators [120]. δ¹⁸O analyses were performed using a TC-EA (high-temperature conversion-elemental analyzer) coupled online to a ThermoFinnigan Delta V Plus mass spectrometer. 0.2 to 0.3 mg Ag₃PO₄ was weighed into silver foil and transferred to the sample carousel of the TC-EA. At 1450°C, the silver phosphate was reduced, and CO formed as the analyte gas [121]. CO was transferred in a helium stream through a gas chromatograph via a Conflo III interface to the mass spectrometer. Samples as well as standards were generally measured in triplicate. The measurements were calibrated by performing a two-point calibration [122] using NBS 120c (21.7 ‰) and a commercial Ag₃PO₄ (9.9 ‰). A laboratory standard was used as a control standard and processed together with the samples. All standards were calibrated to TU1 (21.11 ‰) and TU2 (5.45 ‰) [121]. External reproducibility, monitored by replicate analyses of samples as well as the laboratory standard was ± 0.11 ‰ (1σ; n = 3). The average oxygen isotope composition of the internationally distributed standard NBS 120c was measured as 21.70 ± 0.06 ‰ (1σ; n = 8).

Gustorzyn

The ⁸⁷Sr/⁸⁶Sr values of the human enamel vary within a relatively wide range from 0.7098 to 0.7125 (Tables 1 and 2; Fig 3). However, there is a single outlier in the data set–a female aged 20–30 from grave 5 (MAK56), who exhibits the most unradiogenic Sr isotope signature. Discounting this value (0.7098), the population at Gustorzyn shows in fact a narrow range, from 0.7112 to 0.7125, and there are no significant differences in the Sr isotope composition of males (0.7112–0.7123, n = 7) and females (0.7115–0.7119, n = 3). Due to the small number of samples, it is difficult to say whether this feature was also characteristic of the entire population of Gustorzyn. The range of values measured for juveniles of undetermined gender (0.7113–0.7125) only slightly exceeds the range of males.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Strontium (⁸⁷Sr/⁸⁶Sr) and oxygen (δ¹⁸O_P-VSMOW) isotope composition in tooth enamel of humans from cemeteries in Gustorzyn and Żerniki Górne.

https://doi.org/10.1371/journal.pone.0282472.g003

δ¹⁸O_P ratios were measured for 15 out of 17 samples subjected to ⁸⁷Sr/⁸⁶Sr analyses. The δ¹⁸O_P results range from 15.9 ‰ to 18.3 ‰ and have a normal distribution (Shapiro-Wilk test, W = 0.900, Two-tailed p-value = 0.115, α = 0.050). However, two individuals have slightly lower δ¹⁸O_P ratios. First (15.9 ‰) was obtained for the above-mentioned female from grave no. 5 and the second (16.0 ‰) was measured for the child (individual 5, aged 5–6) from the same feature. In the case of other individuals, there are no clear differences in δ¹⁸O_P values related to age and/or sex. In case of juveniles, δ¹⁸O_P values range from 16.9 ‰ to 18.3 ‰ (n = 3), for males it is a range from 17.0 ‰ to 18.3 ‰ (n = 7), and for females between 17.1 ‰ and 17.8 ‰ (n = 3).

Żerniki Górne

⁸⁷Sr/⁸⁶Sr and δ¹⁸O_P isotope signatures were measured for enamel samples taken from all 18 individuals. The ⁸⁷Sr/⁸⁶Sr values vary within a very narrow range from 0.7082 to 0.7091 and there is no correlation between the values obtained for the examined individuals and their age and/or sex (Table 2; Fig 3). ⁸⁷Sr/⁸⁶Sr ranges for females and males are almost identical, from 0.7082 to 0.7091 (n = 8), and from 0.7082 and 0.7088 (n = 8), respectively.

δ¹⁸O_P values range from 15.3 ‰ to 16.5 ‰ and have a normal distribution (Shapiro-Wilk test, W = 0.949, Two-tailed p-value = 0.449, alpha = 0.050). In case of females, the range is 15.7–16.3 ‰ (n = 8), for males 15.3–16.5 ‰ (n = 8), and for children 16.2–16.4 ‰ (n = 2). Hence, there is no correlation between the measured δ¹⁸O_P values and the sex and/or age of studied individuals.

Sr isotope background of Kuyavia

Located in north-central Poland, Kuyavia is a relatively small region (ca. 8,000 km²) with vague borders defined on historical and ethnographic bases. Kuyavia consists of five main sub-regions (the Kuyavia Plain, Kuyavia and Żnin Lakelands, Toruń and Płock Basins) and is covered by Pleistocene glacial sediments of the Vistulian ice sheet and alluvial deposits of Holocene age [125, 126] (Fig 4). The glacial sediments are highly variable in lithology from clays to coarse-grained deposits and tills. They all consist of allochthonous clastic material, derived entirely from Scandinavia. This material is a product of glacial erosion and weathering of Precambrian igneous and metamorphic rocks of the Scandinavian Shield and, to a lesser extent, Palaeozoic and Mesozoic sedimentary rocks. The dominance of material from acidic igneous and metamorphic rocks is the reason for the high (radiogenic) ⁸⁷Sr/⁸⁶Sr ratios of the Pleistocene bedrock of the Kuyavia region, analogous to the ratios (average 0.730) observed in Precambrian rocks in the Fennoscandia region [127, 128]. Glacial sediments are thus the most radiogenic element of the local environment, with ⁸⁷Sr/⁸⁶Sr values everywhere above 0.72 (for review see [68]). The Kuyavia Plain, where Gustorzyn is located, consists of tills, weathered tills, and glacial sands and gravels.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Geological map of Kuyavia with location of the Gustorzyn site, other cemeteries and settlements of the Trzciniec cultural circle societies, and other sites mentioned in this study.

The map was produced with the QGIS 3.26.0 software, based on data from the Geological Map of Poland [129], and on data downloaded from https://www.hydrosheds.org/products/hydrorivers (License Agreement) and https://geoportal.gov.pl. Borders of Kuyavia and at the same time the area systematically surveyed by archaeologists is marked by grey dashed line.

https://doi.org/10.1371/journal.pone.0282472.g004

The Kuyavia Plain is an undulating ground moraine, with melt-out depressions, ice-marginal valleys and subglacial channels. The latter are filled with lakes, of which Gopło is the largest. The biggest river is Noteć, followed by Parchania and Bachorza and their tributaries: Tążyna and Zgłowiączka (Fig 4), respectively. In contrast to the geological substrate, the riverine water of the Noteć and its tributaries exhibits moderately radiogenic ⁸⁷Sr/⁸⁶Sr values, lowered by inputs of chemical fertilizers in agriculture and mine waters from an open-pit lignite mine [73]. In the southern part of the region, where the Noteć River in its upper course is less affected by anthropogenic activities, the ⁸⁷Sr/⁸⁶Sr ratio has a maximum value of 0.7127. Groundwater from two natural outflows in the Żnin Lakeland (west of the Kuyavia Plain) yielded values of 0.7116 and 0.7111 [130]. It can be assumed that in the historical times the Sr isotope composition of groundwater could have been at this level or even was slightly more radiogenic.

A characteristic feature of the Kuyavian geology is the presence of Permian salt diapirs that pierced the Mesozoic strata and thus in some places occur directly under the Quaternary cover [131]. They constitute sources of ascending brines that influence locally the composition of groundwaters. This is because the brines are characterized by an unradiogenic composition with ⁸⁷Sr/⁸⁶Sr values below 0.7091 [73]. However, the main source of unradiogenic Sr in Kuyavia is atmospheric water (rain and snow) which has ⁸⁷Sr/⁸⁶Sr value of ~0.7092 identical to that of seawater [132, 133].

Sr isotope data for archaeological humans and animals from Kuyavia

A relatively large pool of comparative data is available for the Kuyavia region in terms of variability of ⁸⁷Sr/⁸⁶Sr ratios in skeletal remains of prehistoric and early historic humans and animals. For three humans from a Late Neolithic population from Kierzkowo in the Żnin Lakeland the range of ⁸⁷Sr/⁸⁶Sr values measured in 11 teeth ranged from 0.7117 to 0.7140 but the range for local humans probably did not exceed 0.7135 [130]. In the case of a Roman period population from the cemetery in Karczyn-Witowy (western Kuyavia Plain), possible non-local individuals (two out of 15) showed ⁸⁷Sr/⁸⁶Sr values of 0.7092 and 0.7131, while others exhibited values between 0.7104 and 0.7123 [134]. The next two sites are located in the eastern part of the Plain, close to Gustorzyn. For a single Final Neolithic burial in Kruszyn two enamel samples yielded values of 0.7110 and 0.7123 [41]. Much lower values were reported for an early medieval population from the cemetery in Bodzia. Most individuals (11 out of 12) yielded values from 0.7090 to 0.7118 interpreted as non-local and only the male with the ⁸⁷Sr/⁸⁶Sr signature of 0.7129 was considered to be local [37, 135].

Enamel samples of fauna from sites in the Kuyavia Plain (Karczyn, Kruszyn, Ludwinowo; n = 3) exhibit values in the range 0.7124–0.7126 [134], at a number of sites from eastern part of the Plain (Bodzia, Ludwinowo, Kruszyn) these were 0.7121–0.7131 [37], in its southern periphery, close to Kuyavia Lakeland, they were between 0.7117 and 0.7130 (n = 9) [136]. In the Żnin Lakeland enamel samples of pig, goat and fox (n = 3) ranged from 0.71296 to 0.71339, while a series of sequential samples along a single cattle tooth (n = 14) gave unradiogenic ratios, from 0.7108 to 0.7113, interpreted as evidence of non-local origins of that specimen [130].

In the light of discussed data, the majority of people buried in Gustorzyn (with ⁸⁷Sr/⁸⁶Sr signatures from 0.7112 to 0.7125) most probably lived within this part of Kuyavia. The listed data also illustrates that ⁸⁷Sr/⁸⁶Sr ratios measured for enamel of archaeofauna from a number of sites in Kuyavia were clearly higher than those measured for humans.

Sr isotope background of the Nida Basin

The cemetery in Żerniki Górne is located in the central part of the Nida Basin which constitutes the marginal part of the Carpathian Foredeep Basin, developed during Miocene times due to thrust loading of the European foreland lithospheric plate by the Carpathian orogeny [137]. The sedimentary infill of the Nida Basin is represented primarily by thick, Miocene age argillaceous deposits (clays) containing layers of evaporates (gypsum). These rocks are widely exposed in the area around the town of Busko Zdrój (Fig 5). They overlay here a Mesozoic substrate composed of the Upper Cretaceous carbonates and marls. Northeast of Żerniki Górne, the Miocene rocks are covered by the Pleistocene deposits. Predominant among these are glacial tills and fluvioglacial sands which are Middle Pleistocene in age. Southeast of the site, the Pleistocene sequence terminates with a continuous cover of the Vistulian loess deposits [71, 138].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Geological map of southern Nida Basin with location of Żerniki Górne, other cemeteries and settlements of the Trzciniec cultural circle societies, and other sites mentioned in this study.

The map was produced with the QGIS 3.26.0 software, based on data from the Geological Map of Poland [129], and on data downloaded from https://www.hydrosheds.org/products/hydrorivers (License Agreement) and https://geoportal.gov.pl/. Note that area systematically surveyed by archaeologists is marked by grey dashed line and it does not include the surroundings of Żerniki Górne.

https://doi.org/10.1371/journal.pone.0282472.g005

The Sr isotope composition of bedrock in the Żerniki Górne area is relatively well-known. ⁸⁷Sr/⁸⁶Sr values of 0.7088 to 0.7091 were reported for the Miocene gypsum deposits [139]. Recently, [71, 138] provided Sr isotope data for Miocene clays (values from 0.7144 to 0.7152) and Vistulian loess deposits (values from 0.7200 to 0.7240) exposed in the Nida Basin. The Sr isotope composition of the Upper Cretaceous substrate has not been measured so far. However, as these are marine carbonate rocks unradiogenic Sr signatures between 0.7074 and 0.7077 are to be expected [132]. Hence, around the investigated site at Żerniki Górne, an extremely wide variation in Sr isotope composition appears to characterize the local environment, from strongly unradiogenic to highly radiogenic signatures. In contrast to the wide variation in Sr isotope signatures of rocks, the riverine waters exhibit everywhere an unradiogenic composition. Signatures of the Vistula, Szreniawa, and Nida rivers vary within a narrow range between 0.7084 and 0.7095 [68].

Sr isotope data for archaeological humans and animals from the Nida Basin

For the Nida Basin Sr data for humans is available from three sites. The first one is a Late Neolithic mass grave from Koszyce [71, 138]. It contained the remains of 15 people. ⁸⁷Sr/⁸⁶Sr values measured for all individuals range from 0.70963 to 0.71201, values for three samples of pig tooth enamel are between 0.70948 and 0.71126. It was proposed that humans with low values (0.7105–0.7110) are local, and those with more radiogenic Sr isotope composition (0.7113–0.7120) were born outside Koszyce area.

The next two sites are dated to Final Eneolithic. ⁸⁷Sr/⁸⁶Sr values for three humans from Malżyce, located in a loess upland ranged from 0.70954 to 0.71028 [71]. 13 individuals from Gabułtów exhibit Sr isotope ratios ranging 0.70956 to 0.71039 [71]. This cemetery is located further south from Malżyce, in an area dominated by Miocene clays. Despite differences in geological settings both groups exhibit similar ⁸⁷Sr/⁸⁶Sr values. All humans from both sites were interpreted as being of local origin, which corresponds to the local character of pottery vessels and flint tools found in their graves [71].

The animal baseline samples come from Gabułtów itself (horse tooth, ⁸⁷Sr/⁸⁶Sr value of 0.71026), and to locations in a loess upland, Michałowice (a horse tooth with ⁸⁷Sr/⁸⁶Sr value of 0.71117) and Jakuszowice (various domesticates with ⁸⁷Sr/⁸⁶Sr values ranging from 0.70996 to 0.71233). Due to possible high mobility of horses their values are not especially useful for establishing local baseline values.

The very narrow range of human ⁸⁷Sr/⁸⁶Sr values in Żerniki Górne (0.7082–0.7091) perfectly matches the isotopic composition of the Miocene sediments that form the main element of the local geological substrate. There is no direct information on the Sr isotope composition of surface and/or groundwater waters at Żerniki Górne. However, they are anticipated to be unradiogenic in composition as the nearest large river, Nida, draining a substrate with a similar lithological composition 15 km to the west, displays ⁸⁷Sr/⁸⁶Sr values from 0.7084 to 0.7091 [68]. As the Quaternary cover of radiogenic loess deposits is only rudimentary preserved in the eastern part of the Nida Basin (Fig 5), it can be assumed that the water available to people in Żerniki Górne was even less radiogenic. All of this provides compelling evidence for the local origin of all the individuals studied.

Oxygen isotope baselines and variability across the two study areas

Oxygen isotope in bioapatite is typically used for tracking mobility over longer distances than Sr isotopes and owing to the low number of samples for oxygen isotopes, baselines and variability across both study areas are discussed here.

For Kuyavia δ¹⁸O values measured in various tissues were reported for two Late Neolithic humans from Kruszyn (δ¹⁸OC-VPDB –4.75 ‰, = δ¹⁸O_P-VSMOW 26.01 ‰ = δ¹⁸O_DW –7.27 ‰ [41]) and Kowal (δ¹⁸O_P-VSMOW from 20.86 to 25.23 ‰ = δ¹⁸O_DW from –1.60 to 5.13 ‰ [40]), and above mentioned early medieval population from Bodzia [37]. However, the majority of individuals from the latter site were assessed as being non-local based on their ⁸⁷Sr/⁸⁶Sr signatures so their δ¹⁸O values do not necessarily mirror local δ¹⁸O precipitation values.

δ¹⁸O_C-VPDB values have also been measured in tooth enamel of sheep, goat and sheep/goat from the Middle-Neolithic settlement in Kopydłowo located in the south of Kuyavia Plain [136]. The measurements were made along the M2 and M3 tooth profiles, which allowed to trace changes in O isotope ratios in the environment in an annual cycle. The obtained values after conversion to δ¹⁸O_P-VSMOW are in a range from 10.22 to 17.54 ‰ = δ¹⁸O_DW from –17.98 to –6.70 ‰. For an Early Neolithic LBK settlement in Ludwinowo (eastern Kuyavia Plain) ten cattle M3 teeth were sequentially sampled. Complete data for these samples has not been published but reported δ¹⁸O_C-VPDB values are between –8.4 and –2.5 ‰ [140], = δ¹⁸O_P-VSMOW from 13.3 to 19.6 ‰ = δ¹⁸O_DW from –13.3 to –3.6 ‰. As expected, O isotope composition in animals is much different than in humans, and values measured for the former cannot serve as a baseline for the latter.

Comparative O isotope data from Nida Basin is limited to above mentioned Final Neolithic cemetery in Malżyce [42] (Table 3). In addition, oxygen isotopes have been measured for humans buried at an early medieval (11^th c. AD) cemetery in Kraków [38], i.e., outside the Nida Basin but close to its western border (Fig 5). Samples of femur were collected for 65 individuals, including 11 children older than 5 years (most likely no longer breastfeeding) and 54 adults (22 males and 32 females).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Annual δ¹⁸O_DW-VSMOW values estimated using Online Isotopes in Precipitation Calculator (OIPC [115]) and δ¹⁸O_P-VSMOW ratios measured for skeletal tissues of humans from prehistoric and early historic sites in Poland and north-eastern Germany discussed in this study (reported as δ¹⁸O_P-VSMOW or calculated from δ¹⁸O_C-VPDB values).

δ¹⁸O_DW values for humans calculated from measured δ¹⁸O_P-VSMOW and δ¹⁸O_C-VPDB values. In case of humans O isotope composition was measured in single samples of enamel, except the individual from Kowal, for whom enamel, dentin and bones samples were analysed. Location of Tollense after [145], simplified.

https://doi.org/10.1371/journal.pone.0282472.t003

To aid in the interpretation of the data and to help estimate the local baselines and variability a precipitation δ¹⁸O isoscape for the study regions (Fig 6) was generated using raster grid from OIPC v3.2 database [115], based on data from the Global Network of Isotopes in Precipitation and method of [141]. We accept that this database uses modern precipitation δ¹⁸O data which may not be strictly comparable to past δ¹⁸O_DW values but nevertheless represents a reasonable place to start estimating past δ¹⁸O_DW. Fig 6 shows a strong Northwest-Southeast gradation in precipitation δ¹⁸O. Despite being nearly 300 km apart in latitude sites in east Kuyavia, near Gustorzyn, and in the Nida Basin have a similar and narrow range in precipitation δ¹⁸O from –9.3 to –9.1 ‰ (Table 3; Fig 6), in the North this trends to lower values westward towards Drawsko [35] and Górzyca [39].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Map of estimated annual δ¹⁸O isotope ratios in precipitation in East-Central Europe.

The map was produced with the QGIS 3.26.0 software, based on data from Online Isotopes in Precipitation Calculator (OIPC) database (v3.2, waterisotopes.org). Measured and/or estimated (in brackets) mean δ¹⁸O_DW values for humans from selected sites based on data from Table 3. Inset shows location of the study area in Europe.

https://doi.org/10.1371/journal.pone.0282472.g006

Using estimated δ¹⁸O_DW values for the site from the model isoscape it is possible to calculate for each site Δ OIPC–bioap δ¹⁸O_DW value which is the difference between the model δ¹⁸O_DW and the measured mean δ¹⁸O_DW for humans, based on converting δ¹⁸O_P or δ¹⁸O_C to δ¹⁸O_DW values using Eqs (3) and (6) respectively. For most sites across the study areas the mean Δ OIPC–bioap δ¹⁸O_DW is <2.3 ‰ which is small in the context of the considerable error (>1 ‰) associated with the conversion of δ¹⁸O_P and δ¹⁸O_C values to δ¹⁸O_DW. For Gustorzyn and Żerniki Górne Δ OIPC–bioap δ¹⁸O_DW value is –2.2 ‰ and –0.3 ‰ respectively, at the nearest Bronze Age site with O isotope data, Górzyca [39] (Fig 7) Δ OIPC–bioap δ¹⁸O_DW is also small at –1.0 ‰ (Table 3). The second Bronze Age population at Tollense, cannot serve as a benchmark here, as many of the studied humans were of non-local origins [14] (Fig 7). In the case of an early modern population from Drawsko [35] (Fig 6) significant processing of drinking water can be expected, leading to a gap between estimated and measured values. Still, this difference is low (–1.7 ‰) (Table 3). Comparative O isotope data in the Nida Basin is limited, however, for the medieval people from Kraków Δ OIPC–bioap δ¹⁸O_DW value is low (1.1 ‰).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Strontium (⁸⁷Sr/⁸⁶Sr) and oxygen (δ¹⁸O_P-VSMOW) isotope composition in human remains from Gustorzyn and Żerniki Górne (this study), and from Middle-Late Bronze Age sites in Górzyca [39] and Tollense [14].

https://doi.org/10.1371/journal.pone.0282472.g007

Across both study regions, two oxygen isotope data sets standout as having very high Δ OIPC–bioap δ¹⁸O_DW values and are therefore in poor agreement with the modelled δ¹⁸O_DW values, these are the individual from Kowal in Kuyavia and the individuals from Malżyce in the Nida Basin.

For the individual from Kowal δ¹⁸O_P values have been measured on enamel, dentine and bone. Considering only the two enamel measurements on the first incisor and premolar, δ¹⁸O_P values are 25.23 and 22.21 ‰ respectively, giving a δ¹⁸O_DW value of –12.1 ‰ which is far outside the range for this part of East-Central Europe (Fig 6). This has already been noted by authors of the Kowal study [40]. Such a large discrepancy is difficult to be explained and the authors seem to be aware of that as they did not explicitly support the local or non-local origin of that individual [40]. Moreover, this single individual exhibits a greater range in oxygen isotope values (SD = 1.9) than all other populations in our study area.

For the site of Malżyce, the 13 individuals measured have mean δ¹⁸O_P values of 20.5 ‰, significantly high when compared to the nearby site of Żerniki Górne with a mean δ¹⁸O_P value of 16.1 ‰. Moreover, their Δ OIPC–bioap δ¹⁸O_DW of –7 ‰ is surprisingly low when compared to other sites, except Kowal (Table 3; Fig 7). What is also clear is the data from the Malżyce site exhibit a significantly greater range of δ¹⁸O_P values (SD = 1.3) (Table 3) compared to the contemporary site of Żerniki Górne (SD = 0.3) and all other sites discussed here (SD between 0.7 and 0.8), again except Kowal and Kraków (Table 3). Whilst it cannot be excluded that such a high variability of values for Malżyce reflects highly variable provenance of people buried there, similar cases are unknown to us. Even people recovered from Tollense battlefield exhibited lower variability in O isotope composition (SD = 0.8) [14] compared to Malżyce. This is important as their strontium, carbon (δ¹³C_C) and lead (²⁰⁴Pb, ²⁰⁶Pb, ²⁰⁷Pb, ²⁰⁸Pb) isotope ratios clearly indicate they came from many regions, often very distant from the place of their death [14]. High δ¹⁸O_P variability (SD = 1.4) was also found among humans from Kraków [38].

For Malżyce and Kowal δ¹⁸O_P analyses were carried out for a mixture of enamel, dentin and bone samples, while for Kraków bone samples were analysed exclusively which is a rare case. For instance, a comprehensive meta-study of medieval sites from England and continental Europe [142] brings only one case when oxygen isotopes were measured in bones [143]. Here humans from an early medieval cemetery in Volders (Austria) can be characterised with a wide range of δ¹⁸O_C values between –16 and –5 ‰. After conversion to δ¹⁸O_P they range from 5.3 to 16.9 ‰ with a significantly high SD of 2.1. We are aware of two other studies based on bones. A medieval cemetery in Gródek at today’s Polish-Ukrainian border [36] again brings δ¹⁸O_P data of high variability with SD of 1.2 (Table 3). However, in a study of the early medieval cemetery in Povegliano Veronese in Norther Italy both enamel and bone samples of the same individuals were analysed, and bone samples had a significantly lower range of δ¹⁸O_P values than enamel [144]. Hence, it is currently inconclusive if δ¹⁸O values in bone can be more variable than in enamel. Moreover, we cannot explain why humans from Kowal and Malżyce exhibit much higher δ¹⁸O_P values than both: values expected from the OIPC model, and values measured for humans (both in carbonates and phosphates) from nearby archaeological sites. Hence, hereafter we will not refer to data for Kowal, Malżyce and Kraków (together with Gródek all coming from the same laboratory).

Mobility in the Trzciniec cultural circle societies

The isotopic data from Gustorzyn shows that the majority of studied individuals are consistent with a local origin. The suspected non-local individual is a female aged 20–30 (individual 6 from grave 5). She exhibits outlying ⁸⁷Sr/⁸⁶Sr values (0.7098) and also the lowest δ¹⁸O _P-VSMOW (15.9 ‰ = δ¹⁸O_DW –9.2 ‰) value for the site. Her ⁸⁷Sr/⁸⁶Sr value is lower than the majority of prehistoric and early historic humans from Kuyavia who have been assessed as local in origin. A lower value (0.7092) was noticed only for 16–18 old individual of unknown sex from the Roman Period cemetery in Karczyn-Witowy [134]. Her ⁸⁷Sr/⁸⁶Sr value is consistent with an origin on relatively unradiogenic marine bedrock and with δ¹⁸O_DW values in between those which are characteristic for, for example, Kuyavia and the Nida Basin. One such area is the borderland of Greater Poland and Polish Jura, in the Upper Warta River basin, located up to 150 km S-W from Gustorzyn and known for clusters of TCC sites [45, 146]. δ¹⁸O_DW values for the southern-most cemetery of that cluster, in Strugi, can be estimated as –9.2 ‰ (Fig 6), which well fits the value received for the discussed woman. Here, the bedrock is Pleistocene sediments, locally eroded exposing Middle Jurassic marine rocks. Hence, one can expect relatively unradiogenic ⁸⁷Sr/⁸⁶Sr, which would be consistent with the hypothesis that this was a region where the woman spent her childhood. Alternatively, her Sr and O isotope ratios could be a result of migration from an area located further south, for instance in southern Poland, with lower Sr, although with similar O isotope values. If her journey took place during the period of enamel mineralization as an infant, her isotopic signatures would be a result of mixing the initial Sr and O values with higher values on the way North, although this is speculation. The non-local origin of the women from Gustorzyn may suggest the presence of female exogamy in the studied population. This was also noted by anthropometric analyses of cranial proportions [48, 56], although more isotopic information is needed to determine if this is an exceptional case or a more widespread practice.

The second individual from Gustorzyn with relatively low O isotope ratio is a child (individual 5) with δ¹⁸O_P values of 16.0 ‰ and an ⁸⁷Sr/⁸⁶Sr of 0.711670 which is consistent with a local origin. The interpretation of this individual is complex. There is a distinct possibility that this individual was breastfed, and this should lead to elevated δ¹⁸O_P values compared to the δ¹⁸O_P values expected for local adults i.e., the O isotope composition in local drinking water would have to be even lower than the equivalent of δ¹⁸O_P 16.0 ‰ = δ¹⁸O_DW –9.1 ‰. This would indicate that this child was of non-local origins and came from a region of relatively low δ¹⁸O_DW values, but with similar geology to Kuyavia Plain. Such a scenario is, however, rather unlikely. Britton and collaborators [112] found that infants from a medieval cemetery at Wharram Percy, England, exhibited elevated δ¹⁸O_P values in their deciduous teeth. However, already after the age of three, δ¹⁸O_P values measured in bones were >1 ‰ lower, similar to adult mean values or even lower than the equivalent local δ¹⁸O_DW values at the age 4–5 [112].

It is a challenge to assess what is the rate of mobility of the “local” people from Gustorzyn. Their ⁸⁷Sr/⁸⁶Sr values indicate that they had to have spent their childhood in an area covered by glacial sediments, which cover a large part not only of Kuyavia but also most of today’s northern Poland [129]. However, we observed that humans and animals from the flat moraine of the Kuyavia Plain exhibit distinctly less radiogenic Sr isotope ratios than majority humans from neighbouring regions, where ⁸⁷Sr/⁸⁶Sr values might exceed 0.7135 [37, 147]. Hence, we can suppose that people buried at Gustorzyn most probably limited their activity to eastern Kuyavia. In such a case they could exploit the area immediately surrounding the cemetery, and settle along Zgłowiączka, the nearest larger source of drinking water. Hence, the isotopic data is in agreement with the site-catchment model, assuming human activity within a radius of 2 km from the central settlement and nearby cemetery.

δ¹⁸O_P signatures (16–18.3 ‰) of people from Gustorzyn are not very helpful in assessing their short-range mobility, as similar values (15.7–18.9 ‰) were received for a cemetery in Drawsko [35], located as far as 190 km to the West (Fig 6). Next cemetery with O data, Górzyca, is located 280 km West from Gustorzyn, and also has similar values from 16.3 to 18.9 ‰ [39] (Table 3). In other words, the variability of O isotopes in this part of Europe is insufficient to trace journeys on the East-West axis shorter than at least 300 km. Travels along North-South axis are easier to detect, as δ¹⁸O_DW values are more variable due to larger changes in altitude and distance from the sea. Moreover, the equipment of both graves at Gustorzyn does not include any non-local objects which could indicate that one of the buried humans came from regions outside the TCC-world. The most distinctive, cultural tradition contemporary with the “Trzciniec” was the Tumulus culture, covering regions located further West and South-West, the nearest being Greater Poland and Silesia [146, 148]. The stylistics of bronze items from Gustorzyn resemble the Tumulus culture style but such southern traits are common in Kuyavia and point towards intensive interactions with the neighborhood rather than migration(s) [148]. They were preceded by trans-Carpathian influences from the Otomani-Füzesabony circle, though they were less intense here than in the Nida Basin [149].

In contrast to Gustorzyn, it is highly probable that all studied individuals from Żerniki Górne are of local origin and spent their infancy close to the place where they were ultimately buried. Some of the communal graves at this site contain equipment that might evidence influences of other cultural groups [51]. This is a broader regularity, also visible at other sites in the uplands of southern Poland and associated with intense contacts between local elites and highly developed communities of the Otomani-Füzesabony culture from the Carpathian Basin and its northern edges [29, 149–153]. The style of pottery vessels (types of containers, specific ornaments) and bronze items were clearly inspired by “Southern” patterns. Although the exogamy between the TCC populations from Nida Basin and the Otomani-Füzesabony culture communities settled in the Subcarpathian zone cannot be ruled out, the most reliable interpretation of similarities in material culture is far-reaching exchange of amber and metal [21, 22, 29, 153].

The cemetery in Żerniki Górne is located at the edge of two areas with distinct lithologies, with Pleistocene loess prevailing in the East, and Miocene clays and evaporates covering lower parts of the landscape within a radius of at least 5 km around the site (Fig 5). As all humans exhibit unradiogenic ⁸⁷Sr/⁸⁶Sr ratios, which are typical for the Miocene geology, these people most probably occupied these lower areas. It can be hypothesised that they settled at Nida River, flowing 15 km South-east from the site, and draining non-radiogenic Cretaceous bedrock. The latter would be the most plausible source of water with unradiogenic Sr. This can also be seen the other way round: Sr isotope composition in human tooth enamel allows to predict the location of an area exploited by these humans. This stays in agreement with archaeological observation of settlements being located along rivers, and cemeteries occupying higher parts of the landscape (Fig 5).

The range of ⁸⁷Sr/⁸⁶Sr values for humans from Żerniki Górne is extremely narrow when compared to other human populations, both in Nida Basin and Kuyavia. This might indicate that the area they exploited for farming and husbandry was geologically homogeneous. However, considering the relatively high variability of lithology in radius of ~10 km around the site, their mobility was most probably limited to very specific and rather small parts of the landscape. As in case of Gustorzyn, O isotope data is not helpful in testing short-distance mobility of the studied humans. Nevertheless, it allows to reject the possibility of long-distance travels from regions with similar geological settings as Żerniki Górne which would lead to ⁸⁷Sr/⁸⁶Sr ratios indistinguishable from the local signal.