The Roman Military Community as a Melting Pot: Biomolecular Evidence from the Lower Rhine Limes

The Roman Military Community as a Melting Pot: Biomolecular Evidence from the Lower Rhine Limes | Research Square Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Roman Military Community as a Melting Pot: Biomolecular Evidence from the Lower Rhine Limes Maura De Coster, Fabricio Furni, Wouter K. Vos, Jeroen Oosterbaan, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8699464/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Military communities in the Roman borderlands are increasingly understood as socially mixed and dynamic rather than homogeneous military enclaves. This study adds a biomolecular perspective through combined strontium, oxygen and carbon isotope analysis, and ancient DNA of 33 inhumed and 44 cremated individuals buried at Praetorium Agrippinae in modern-day Valkenburg (South Holland, the Netherlands). The site contains one of the largest, well excavated Roman-period cemeteries in the Lower Rhine Limes Borderlands, dominated by cremations with a smaller yet substantial inhumation component. Sr-O-C isotope data indicate a large variety of geographical regions of origin among both inhumed and cremated individuals, demonstrating sustained mobility into the settlement throughout its 300-year existence. Ancient DNA analysis establishes heterogeneous ancestry profiles and an absence of close kinship ties, further supporting a diverse population with varied geographic origins. Overall, the biomolecular evidence positions the Roman inhabitants of Praetorium Agrippinae as a melting-pot borderland community shaped by movement, demographic diversity, and social differentiation. Roman borders isotopes aDNA military community Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction In recent decades, the concept of military communities has transformed how scholars interpret life along the Roman borderlands. Earlier limes research, rooted in the traditions of the Reichs-Limeskommission , primarily treated forts, canabae and vici as separate archaeological site types, focusing on architectural and unit histories (Johnson 1983 ; Breeze 1987 ). Since the late 1990s, however, studies have reframed the Roman army as a social formation embedded in wider communities rather than an isolated entity. The seminal volume The Roman Army as a Community (Haynes 1999a ) catalysed this shift by emphasizing households, religion, identity, and interaction between soldiers and civilians. Subsequent work has expanded this perspective through analyses of domestic assemblages, gender, age, and identity within forts and associated settlements (Allison 2013 ; Greene 2019 ; Gardner 2016 ; Collins and McIntosh 2014). As a result, the term military community has become standard in Roman borderland archaeology, capturing the social, economic, and familial entanglements that structured life along the limes . Bioarchaeological approaches can directly address the mobility that structured Roman military communities in borderland contexts by combining strontium, oxygen, and carbon isotope analyses with aDNA to reconstruct childhood origins, migration histories, kinship, and population structure. Multi-isotope studies in northern provinces have demonstrated the power of these methods to identify non-local individuals and trace patterns of recruitment and provisioning (Chenery et al. 2010 ; Madgwick et al. 2019 ), while integrated isotopic and genomic work increasingly reveals how movement and ancestry shaped life at the empire’s margins (Scheib et al. 2024; Depaermentier et al. 2023 ). Despite this potential, bioarchaeological research explicitly targeting military or fort–vicus assemblages in Roman borderlands remains limited. Reviews emphasize that mobility studies have focused mainly on urban or cemetery contexts, with restricted baselines and fragmented datasets constraining broader applications (Shaw et al. 2016 ; Eckardt, Müldner, and Lewis 2014 ). Recent frameworks developed for the Lower Rhine Limes highlight both the feasibility and interpretive power of systematic isotopic sampling in these regions (Kootker, Van der Velde, and Heeren 2022 ; Kootker and Heeren 2022). Expanding such integrated approaches promises to illuminate how movement and interaction shaped the formation, diversity, and cohesion of Roman military communities in the borderlands. In Valkenburg (South Holland, the Netherlands) the Roman military base Praetorium Agrippinae was excavated 45 years ago and has since remained an archaeological enigma. As one of the most north-western military settlement along the European limes, its location supported a strategic hub function within the Roman military network. The site’s wet preservation conditions have resulted in exceptional organic preservation, particularly of wooden, leather, but also metal artefacts. These finds provide insight into the composition of the military community, including artefacts associated with childhood, such as toys and infant feeding vessels (Smits 2006 ). The presence of women and children is further supported by osteoarchaeological evidence (Smits 2006 ; Lonnée and Maat 1998 ; Dierendonck, Hallewas, and Waugh 1993). Valkenburg site has yielded one of the largest assemblages of both cremated (n = 501) and inhumed (n = 124) human remains from the Lower Rhine Limes, and the largest in the Roman Netherlands, representing multiple phases of occupation between the first and third centuries CE (Mols and Polak 2020; Polak and Kooistra 2013 ; Vos and Oosterbaan in prep.). These remains offer rare direct evidence of the people who inhabited and moved through a Roman military landscape that is otherwise known primarily from architectural and artefactual data. Despite variable preservation and contextual resolution, the assemblage constitutes a robust dataset for isotopic and genetic analyses of mobility, ancestry, and kinship in a military borderland setting. The inhumations can be divided into two groups. The first group consists of neonates and infants (< 1 year, n = 80), that were buried in small coffins or baskets and that were occasionally accompanied by small grave goods(Smits 2006 ). The second group comprises adults (20–70 years, n = 31 in total, n = 18 males, n = 4 females, and n = 9 unknown, based on osteology), and children and adolescents (1–20 years, n = 13). They were generally buried without grave goods, aside from occasional pottery, and were sometimes buried in a manner indicative of limited standardisation in funerary practice (Smits 2006 ; Lonnée and Maat 1998 ). Such burial practices, including prone positions and the interment of multiple individuals within a single grave, are frequently documented in Roman cemeteries (Taylor 2008 ; Arcini 2009 ; Milella et al. 2015 ). Even though most of the individuals were cremated, cremation was not a uniform practice. The cremation graves in Valkenburg exhibited great variation in funerary practise, mainly reflected in the archaeological record through differences in grave type (Table 1 ) (Smits 2006 ). Particularly, bustum -type graves (n = 46) were frequently recorded. In this burial practice, cremation of one or more individuals was performed directly above the grave, after which the burnt remains collapsed into the underlying pit (Noy 2000 ; Vos and Oosterbaan in prep.). Although this grave type is often associated with the Roman military and soldiers (Philips and Wilson 2021; Petculescu 2025 ), its exact origin is still debated (Ota 2007 ; Noy 2000 ). Together, the heterogeneity observed in both inhumation and cremation practices raises important questions regarding the composition, social structure, and cultural background of the Valkenburg community. These patterns point to variability in origin, affiliation, and lived experience that cannot be fully resolved through funerary evidence alone and calls for a study capable of accessing individual biographies. By integrating isotopic (Sr-O-C), elemental, and genomic analyses (McColl et al. 2024 ), this study applies a bioarchaeological framework to explore how mobility and ancestry shaped the social composition of a Roman military community in the Lower Rhine borderlands. The combined approach allows for a direct examination of population diversity and movement within and beyond the Roman army, contributing new evidence to broader debates on identity, migration, and connectivity in the Roman world. This study illustrates biomolecular evidence for Roman military communities as biologically diverse and highly mobile populations, demonstrating how individual life histories were embedded within and actively (re)shaped the networks that defined the Empire’s borderlands. Table 1 Variation of burial practises and number of graves found in Roman Valkenburg. Table adapted from Smits ( Smits 2006 ) and supplemented based on Vos & Oosterbaan (Vos and Oosterbaan in prep.). Typology Vos & Oosterbaan Typology Waugh Description Provincial-Roman grave typology N H A Inhumation 124 Subtotal inhumation graves 124 F B Bustum/ Flat graves of the bustum type 64 C C Brandgruben -graves 173 E D Urn graves with pyre debris 51 D D*/F Urn graves/organic urn graves 25 B E Brandschüttungs - graves 17 A G Knochennester (graves with a concentration of clean cremated remains) 35 G J Disturbed/robbed/unknown 96 I L Cleared pyre sites 40 Subtotal cremation graves 501 Total number of graves 625 Materials & Methods In total, 33 individuals from inhumation graves and 43 individuals from cremation graves were sampled, resulting in 43 diaphyseal samples and 34 rib samples. The cremated bone samples were selected on the basis of preservation and the level of burning, ensuring that all of the samples were fully calcined. The inhumation sampling resulted in 64 enamel samples for strontium, oxygen and carbon isotope analysis, distributed over 21 first molars (M1), 24 second molars (M2), and 19 third molars (M3). The graves from neonates were excluded from this study. Sample preparation The full pre-treatment protocol for strontium isotope analysis and strontium concentration measurements on cremated human remains is outlined in De Coster & Kootker (De Coster and Kootker 2024 ) and briefly described here. The samples were first mechanically cleaned with a Proxxon diamond coated ball burr to remove the outer layer and any remaining sediments. A second step consists of transferring the samples to glass vials and leaching the samples with circa 1.0 mL (10:1 ratio) 1M acetic acid (CH 3 COOH) ultrasonically for 3–10 minutes, followed by two Milli-Q™ rinses, and a 10-minute Milli-Q™ ultrasonic wash (Snoeck et al. 2015 ; Snoeck et al. 2018). This was repeated until the Milli-Q™ in the glass vials was clear or showed a white cloudy colour. Subsequently, the samples were left on the hotplate at 50°C to dry overnight, powdered and transferred into 1.5 mL capped glass vials for further analysis. Similarly, the full sampling protocol for sampling enamel is outlined in Kootker & De Coster (Kootker and De Coster 2024 ) and summarized here. First, the outer layer of the enamel was mechanically removed. Next, 1–10 mg of enamel powder was collected in 2 mL glass vials with screw caps for subsampling or pre-weighted, acid-cleaned Eppendorf® centrifuge tubes and weighed, ready for Sr isotope analysis. Approximately 0.3 mg was weighed into exetainer vials for the O-C isotope analysis. Analytical details Sr-O isotope composition The Sr purification process was done in the clean laboratory facility at the Vrije Universiteit Amsterdam. Both cremation and inhumation samples were dissolved in 0.6 mL of 3M HNO 3 and centrifuged. Once dissolved, 0.5 mL was used for column chromatography, while the remaining 0.1 mL was kept for concentration measurements. The columns consisted of in-house made 2 mL pipette-tips columns filled with 80 µL Eichrom Sr resin (100–150 µm mesh) to extract the strontium (De Coster et al. 2024 ). The purified strontium samples were collected in HCl-HNO 3 cleaned 7 mL Savillex® Teflon vials with conical interior and one drop of 0.5% of H 3 PO 4 was added, dried down and nitrated with 14M HNO 3 . The blanks were spiked with an 84 Sr enriched in-house made solution. The purified Sr samples were redissolved in 2 µL 10% HNO 3 and 1 µl was loaded onto single annealed rhenium filaments with 2 µL TaCl 5 . A Thermo Scientific™ Triton Plus™ thermal ionisation mass spectrometer (TIMS) at the Vrije Universiteit Amsterdam was used to determine the isotope compositions using a static routine. All Sr isotope ratios were corrected for mass fractionation to 86 Sr/ 88 Sr = 0.1194. During this study (2022–2024), the reproducibility of the TIMS was 0.710259 ± 0.000017 for repeated analysis of NIST® SRM® 987 (n = 267, 2s, loading size 200 ng). This is comparable to the long-term reproducibility (2017–2024): 0.710256 ± 0.000018 (n = 696, 2s). NIST® SRM®1400 and an in-house tooth standard were also continually analysed during this study and resulted in a reproducibility of respectively 0.713108 ± 0.000028 (n = 20, 2s) and 0.707864 ± 0.00008 (n = 25, 2s). The column blanks (n = 17) contained an average of 28 pg Sr. This amount is considered negligible when compared to the average amount of strontium present in human enamel and cremated bone.(Harvig et al. 2014 ; Kootker, Van der Velde, and Heeren 2022 ; Snoeck et al. 2015 ; Snoeck et al. 2018). Approximately 0.3 mg of enamel sample was collected in exetainer vials and transferred to the stable isotope laboratory at the Vrije Universiteit Amsterdam for carbon-oxygen isotope analysis. There, the samples were flushed with pure helium (5.0). Phosphoric acid (H 3 PO 4 ) was added for 24 hours at constant temperature of 45°C to dissolve the samples. A Thermo Finnigan GasBench II preparation device, interfaced with a Thermo Finnigan Delta+ mass spectrometer was used to analyse the isotopic composition. The measured isotopes were normalised to the Vienna Peedee Belemnite (VPDB) scale using an in-house carbonate reference material (VICS) calibrated against NBS19 and LSVEC certified reference materials. [Sr], [Ca] and [Pb] measurements Depending on the sample size, aliquots were taken from the remaining 0.1 ml 3M HNO 3 solution and diluted to a total dilution factor of 10,000 in acid-cleaned ICP vials. The strontium ([Sr]), calcium ([Ca]) and lead ([Pb]) concentrations were measured on a ThermoFisher iCAP TQ ICP-MS. Sr (m/z = 88) was measured no gas in the collision-reaction cell, while kinetic-energy-discrimination (KED) using helium (He) was applied for Ca (m/z = 44). Indium (In) was used as an internal standard to correct for sensitivity drift, and quantification was achieved via external calibration against calibrants gravimetrically prepared from certified single-element standards. Repeated digestions and measurements of the NIST® SRM® 1400 (n = 12) yielded recoveries of 100.1-103.2% for Ca and 97.1–98.3% for Sr, with intermediate precision (i.e., repeatability of individual analyses) better than 5% relative standard deviation (1s). All [Sr] from the cremated bone samples were normalised to 40% Ca to account for organic loss using the following equation: $$\:{\left[Sr\right]}_{normalised}=\frac{{\left[Sr\right]}_{measured}}{{\left[Ca\right]}_{measured}}\:\times\:\:40\%$$ DNA Analyses A total of 30 Valkenburg individual whole-genome sequences (individual information and accession codes available in the Supplementary Table 1) were obtained from McColl et al .(McColl et al. 2024 ). Read alignment and ancient DNA authentication were also performed following the procedures described in McColl et al. (McColl et al. 2024 ). Single-nucleotide polymorphism (SNP) genotype calling was initially conducted using BCFTOOLS v. 1.9 mpileup and call(Danecek et al. 2011), applying a minimum mapping quality threshold of > 30 (parameters: -E -q 30 -Q 30). Genome imputation Genomic imputation was performed using the GLIMPSE2 v. 2.0 framework (Rubinacci et al. 2023). High-coverage modern genomes from the 1000 Genomes Project Phase 3 dataset (Siva 2008 ) were used as the imputation reference panel. Only genomes with coverage above 0.4 were selected for imputation (27 individuals). Following imputation, genotypes were filtered based on posterior genotype probabilities, where individuals were retained if the mean posterior probabilities were ≥ 0.95 (26 individuals). After imputation, variant coordinates were lifted to the hg19 human reference genome (GRCh37 (Auton et al. 2015)) using the liftover tool implemented in CrossMap v. 0.7 (Zhao et al. 2014 ). Clustering analyses PCA was performed by projecting the ancient individuals onto present-day populations using the SNP positions from the Affymetrix Human Origins (HO) array (Patterson et al. 2012 ). Genotype data were obtained from the Allen Ancient DNA Resource (AADR) v.54.1 (Mallick et al. 2024). A total of 2,207 present-day Eurasian individuals were extracted from this dataset. Imputed genotypes were merged on 519,224 HO SNP sites using the mergeit utility in EIGENSOFT v8.0 (Patterson, Price, and Reich 2006 ; Price et al. 2006 ). Principal components were estimated using the smartpca module of EIGENSOFT, with ancient individuals projected onto the variation of present-day populations (parameter: lsqproject = YES; numoutevec = 4). Genealogical relationships To investigate genealogical relationships through shared IBD fragments (Ringbauer et al. 2024 ) among contemporary individuals, genomic data for previously sequenced ancient individuals were retrieved from the AADR v.54.1 (Mallick et al. 2024). Additional individuals were obtained from recent publications (Scheib et al. 2024; Antonio et al. 2019 ). A total of 711 individuals from Europe dating within ± 1000 years of the mean date of the Valkenburg individuals (1800 BP) were selected, and the corresponding alignment files were downloaded from the European Nucleotide Archive (individual information and accession codes available in the Supplementary Table 1). Genotype calling and imputation for these individuals were performed following the procedures described above. For IBD analyses, we restricted the data to the 1240K SNP panel(Haak et al. 2015) as these positions are widely represented among previously published ancient genomes. Only 638 individuals with a ≥ 700,000 SNPs covered on the 1240K, and mean genotype posterior probability > 0.95 were retained for downstream analyses. IBD estimation was conducted using ancIBD v0.7 (Ringbauer et al. 2024 ). Phenotype predictions Pigmentation was inferred using the HIrisPlex-S panel for skin, hair, and eye colour (Chaitanya et al. 2018 ; Walsh et al. 2013 ). We applied the probabilistic framework implemented in Phenotypic Inference (Perretti et al. 2025 ), which estimates trait probabilities using genotype posterior probabilities, thereby improving prediction accuracy for low-coverage genomes. Alignment (BAM) files from 28 individuals (> 0.4X) were used as input to derive genotype likelihoods and obtain HIrisPlex-S probability estimates, performing 1,000 replicate inferences per individual. Trait probabilities for each replicate were calculated using the HIrisPlex-S web server (Chaitanya et al. 2018 ; Walsh et al. 2013 ). Final phenotype likelihoods were summarized as percentage support for each predicted trait category based on the HIrisPlex-S posterior distributions. Results Sr-O-C isotopes and Sr concentrations [Sr] Strontium concentrations [Sr], strontium (⁸⁷Sr/⁸⁶Sr), oxygen (δ¹⁸O PDB ), and carbon (δ¹³C VPDB ) isotope ratios were measured in enamel from first (M1), second (M2), and third molars (M3), allowing both inter- and intra-individual comparisons. The descriptive statistics of the Sr-O-C isotope data of the inhumations are shown in Table 2 and depicted in Figs. 1 and 2 . The ⁸⁷Sr/⁸⁶Sr span a broad range across all molars (0.70856–0.71285). Means are similar between molar types (M1: 0.71004; M2: 0.70987; M3: 0.70978), with medians clustering around 0.7096–0.7099. When all molars are combined (n = 64), the mean ⁸⁷Sr/⁸⁶Sr is 0.70982, with ratios distributed on both sides of the established local range (Kootker et al. 2016 ). The δ ¹⁸O PDB range from − 8.3 to − 3.8‰ across all molars. Means are comparable between molar types (M1: −6.1‰; M2: −6.2‰; M3: −6.2‰), with medians between − 5.7 and − 6.2‰. The combined enamel dataset (n = 59) has a mean δ¹⁸O PDB of − 6.1‰, overlapping but also extending beyond the locally reported range (− 6.5 to − 4‰) (Kootker and Heeren 2022; Kootker, Van der Velde, and Heeren 2022 ). Modern dental enamel show intra-dental variation of approximately 0.0002 in ⁸⁷Sr/⁸⁶Sr (Plomp et al. 2020 ) and around 2‰ in δ ¹⁸O PDB (Plomp et al. 2020 ; Kootker, Van der Velde, and Heeren 2022 ). Therefore, observed variation must exceed these thresholds to be considered indicative of mobility. In this study, 45% of individuals displayed intra-molar variation greater than 0.0002 in ⁸⁷Sr/⁸⁶Sr (CL002, CL003, CL005, CL006, CL007, CL023, CL036, CL038, CL041, CL043, CL045, CL046, CL048, CL049), and 58% showed variation greater than 2‰ in δ ¹⁸O PDB (CL002, CL003, CL005, CL006, CL007, CL008, CL015, CL019, CL023, CL024, CL0035, CL036, CL038, CL039, CL041, CL046, CL048, CL049). Notably, 39% of individuals exceeded the mobility thresholds in both isotope systems (CL002, CL003, CL005, CL006, CL007, CL023, CL036, CL038, CL041, CL046, CL048, CL049). The δ ¹³C VPDB show a total range from − 14.9 to − 10.5‰. Means increase slightly from M1 (− 13.0‰) to M3 (− 12.0‰), with corresponding medians of − 12.7, − 12.7, and − 11.8‰ respectively. Interquartile ranges vary between 0.75 and 1.7‰. When combined (n = 59), the δ¹³C VPDB fall largely within the expected range for C₃-based terrestrial diets reported for the region (− 14 to − 11‰), though several samples approach the upper limit of this range. Strontium concentrations ([Sr]) vary substantially between individuals and molar types, ranging from 62 to 266 ppm. Mean concentrations increase from M1 (116 ppm) to M2 (135 ppm) and decrease slightly in M3 (126 ppm). The combined dataset (n = 34) has a mean [Sr] concentration of 125 ppm and an interquartile range of 62 ppm, indicating considerable variability in enamel strontium content. Table 2 Descriptive analyses of the 87 Sr/ 86 Sr,δ 18 O PDB , and δ 13 C VPDB measurements of the inhumations of Valkenburg. Key: M1 – first molar; M2 – second molar; M3 – third molar. 87 Sr/ 86 Sr δ 18 O PDB (‰) δ 13 C VPDB (‰) [Sr] in ppm M1 M2 M3 All M1 M2 M3 All M1 M2 M3 All M1 M2 M3 All Min 0.70884 0.70856 0.70869 0.70856 -8.3 -7.9 -7.8 -8.3 -14.9 -14.9 -13.0 -14.9 62 65 81 62 Max 0.71285 0.71198 0.71225 0.71285 -4.6 -3.8 -5.7 -3.8 -11.5 -10.5 -10.9 -10.5 205 266 176 266 Mean 0.71004 0.70987 0.70978 0.70982 -6.1 -6.2 -6.2 -6.1 -13.0 -12.8 -12.0 -12.7 116 135 126 125 Median 0.70987 0.70954 0.70972 0.70955 -5.7 -6.2 -6.2 -6.2 -12.7 -12.7 -11.8 -12.5 109 149 127 121 Q1 0.70907 0.70912 0.70908 0.70907 -6.7 -6.7 -6.5 -6.7 -13.7 -13.8 -12.5 -13.3 98 85 97 90 Q3 0.71073 0.71011 0.71003 0.71021 -5.3 -5.8 -5.7 -5.5 -12.4 -12.0 -11.7 -11.8 128 158 151 152 IQR 0.00166 0.00099 0.00095 0.00113 1.5 1.1 0.9 1.2 1.2 1.7 0.8 1.5 30 73 55 62 n 21 24 19 64 19 24 16 59 19 24 16 59 12 11 11 34 Local baselines( Kootker et al. 2016 ; Kootker, Van der Velde, and Heeren 2022 ) 0.7088–0.7092 -6.5 to -4 -14 to -11 NA Strontium isotope ratios were measured in calcined bone from diaphyseal and rib samples, alongside normalized strontium concentrations, which can give insights to possible intra-individual mobility. Although turnover rates vary between individuals and are dependent on many variables, generally diaphyseal samples reflect a longer period during life, compared to the ribs that roughly represent the last five years before death (De Coster et al. 2024 ). The ⁸⁷Sr/⁸⁶Sr show a limited overall range across all cremated bone samples (0.70838–0.71121). Means are nearly identical between diaphyseal (0.70954) and rib samples (0.70951), with comparable medians (~ 0.7094). When combined (n = 80), the overall mean ⁸⁷Sr/⁸⁶Sr is 0.70953, with ratios spanning both sides of the locally established baseline range (Kootker et al. 2016 ). Normalized strontium concentrations range from 94 to 227 ppm across all samples. Means are similar between diaphyseal (165 ppm) and rib samples (164 ppm), with medians of 160 and 163 ppm, respectively. In the combined dataset (n = 65), the mean normalized strontium concentration is 165 ppm. Table 3 Descriptive statistics of the 87 Sr/ 86 Sr and normalized [Sr] data of the cremations from Valkenburg Marktveld. 87 Sr/ 86 Sr Normalized [Sr] in ppm Diaphysis Rib All Diaphysis Rib All Min 0.70838 0.70853 0.70838 117 94 94 Max 0.71080 0.71121 0.71121 227 227 227 Mean 0.70954 0.70951 0.70953 165 164 165 Median 0.70940 0.70939 0.70940 160 163 162 Q1 0.70918 0.70908 0.70915 145 136 144 Q3 0.70993 0.70971 0.70988 183 186 183 IQR 0.00075 0.00064 0.00073 38 50 39 n 43 34 77 35 30 65 Sample sizes vary due to preservation and analytical constraints. For inhumations, enamel analyses yielded ⁸⁷Sr/⁸⁶Sr for 64 molars (M1: n = 21; M2: n = 24; M3: n = 19), δ ¹⁸O PDB and δ ¹³C VPDB for 59 samples, and strontium concentrations for 34 samples. For cremations, ⁸⁷Sr/⁸⁶Sr were obtained from 80 calcined bone samples (46 diaphyseal, 34 rib), and normalized strontium concentrations from 65 samples (35 diaphyseal, 30 rib). DNA traces of Valkenburg The principal component analyses (PCA) performed on the DNA data in this study support a heterogeneity in genetic ancestries. While most Valkenburg individuals cluster with present-day Western European populations, notable outliers were present (Fig. 4 A). A 19–28-year-old male (CL032) grouped closely with present-day Greek populations, while a 13–16-year-old male (CL043) clustered with contemporary Estonian individuals. Overall subtle regional differences were recorded. For instance, 16 Valkenburg individuals fell closest to modern populations from the British Isles, whereas seven clustered nearer to populations from France and Spain. Sex ratio and lack of genealogical relationships The molecular sex ratio of the individuals analysed was biased toward males (XY), comprising of 22 males and seven females. Our identical-by-descent (IBD) analyses (Ringbauer et al. 2024 ), which indirectly informs about genealogical affinities, revealed no detectable relationships up to the fifth degree among the Valkenburg individuals. No evidence of such relationships was found between these individuals and more than 600 previously published contemporaneous genomes from the period 1000 BCE to 1000 CE that were previously accessed across Europe (a full list of individuals used is provided in Supplementary Table 1). Valkenburg phenotypes We performed phenotypic reconstructions using the Hirplex-S for the 27 Valkenburg individuals, predicted hair, eye, and skin pigmentation (Chaitanya et al. 2018 ; Walsh et al. 2013 ). The adult male clustering within Mediterranean populations (CL032) showed higher probabilities for an intermediate skin tone, together with dark eyes and brown hair. In contrast, the adolescent male clustering within Baltic populations (CL043) was predicted to have pale skin, blue eyes, and blond hair. Individuals within the western European populations generally exhibited likely intermediate skin, brown eyes, and brown hair pigmentation, although additional phenotypes were present, including red hair and pale skin (Fig. 4 C). Discussion The broad concept of how a military community functioned as a socially and economically fluid entity has relied on historical and material sources. To complement these historically and materially derived characterizations, the isotopic and genetic evidence from Roman Valkenburg offers a biomolecular viewpoint on community composition and mobility. Along the Lower Rhine Limes, the Roman period is characterized by sustained interaction between incoming populations and regional communities (Verhagen, Joyce, and Groenhuijzen 2019 ). In their study, McColl et al. (McColl et al. 2024 ) identified a clear shift from predominantly local ancestry to a more diverse genetic profile during the early Roman period, particularly among individuals from Valkenburg. Similar patterns are reported at other limes sites, where ancestry components indicate movements from eastern and northern regions into late Roman settlements in Germania Superior (Blöcher et al. 2025 ). In contrast, Scheib et al. (Scheib et al. 2024) found that Roman colonization exerted only limited genetic influence on rural populations in Britain. Instead, higher mobility and genetic heterogeneity appear to be concentrated at key nodes of the Roman infrastructural network, especially military and urban centres (Scheib et al. 2024; Olalde et al. 2023 ; Antonio et al. 2019 ), a pattern consistent with the biomolecular evidence from Valkenburg presented in this study that shows pronounced mobility and population heterogeneity within a military-linked settlement. The Sr-O-C isotopic data from the Valkenburg inhumations display an exceptionally wide range of ratios compared to the relatively narrow isotopic ranges observed in the Netherlands (Kootker et al. 2016 ; Kootker, Van der Velde, and Heeren 2022 ; Kootker and Heeren 2022) and surrounding regions (Evans et al. 2023 ; Willmes et al. 2018; Sengeløv et al. 2025 ). The observed variability of individuals also demonstrates mobility within an individual’s lifetime. While minor isotopic variation can arise from physiological or methodological factors, the extent of variation observed in almost half of individuals is more consistent with changes in geographic residence during enamel formation. Particularly notable is the isotopic variation in both strontium and oxygen isotope systems, with nearly two-fifths of individuals exceeding mobility thresholds in both proxies. This covariation strengthens the interpretation of genuine mobility, as these isotopes reflect different aspects of local environmental signatures, namely the local hydrological cycle and food intake. In contrast, the carbon isotopes show a comparatively restricted distribution that is consistent with broadly similar terrestrial dietary practices across the population. Although some individuals have C isotope ratios towards the upper end of the expected regional range (Kootker, Van der Velde, and Heeren 2022 ), overall variability remains limited. This suggests that dietary differences were relatively minor and unlikely to be the primary driver of the pronounced isotopic variability observed in the strontium and oxygen systems. This pattern is consistent with a broadly homogeneous diet dominated by C₃ plant resources, such as cereals and vegetables, in which variations in the relative contribution of animal protein would have had only a limited impact on δ¹³C values. The combination of stable carbon isotope signatures with highly variable strontium and oxygen ratios therefore supports the interpretation that geographic mobility, rather than dietary heterogeneity, underpins the observed isotopic diversity within the Valkenburg population. Although the individual isotope ratios observed among the Valkenburg individuals are compatible with a wide range of European regions, the combined Sr–O isotope data allows assessments of potential regions of origin. One source region frequently suggested in both historical and archaeological interpretations of Valkenburg is the British Isles, owing to the site’s geographical proximity and historical accounts of Caligula’s early attempt to conquer the islands (Vos et al. 2021; Hoffmann 2013 ; Wynia 1999 ; Peddie 2005 ; Manley 2007 ). While our demographic results indicate affinities with present-day British populations, this genetic link may have been established well before the Roman Period (Olalde et al. 2023 ; Cassidy et al. 2025). Previous DNA studies have shown that continental influx of Bell Beaker–related ancestry into the British Isles dates at least from the Copper-Age (Olalde et al. 2023 ). McColl et al. (McColl et al. 2024 ) further demonstrate that the major Valkenburg ancestral component likely derives from a North Sea Bell Beaker origin, which offers a plausible explanation for more distal affinities between the Valkenburg individuals and the modern United Kingdom. However, the combined 87 Sr/ 86 Sr and δ ¹⁸O PDB isotopic data effectively rule out the British Isles as the childhood origin for most of the inhumed individuals. For most individuals, the δ ¹⁸O PDB values are lower than those typically observed and reported in the United Kingdom, suggesting childhood residence in regions less influenced by a maritime climate. Only six individuals (CL005, CL007, CL008, CL019, CL026, CL035) exhibit Sr-O isotope ratios that are compatible with ranges reported for the United Kingdom (Evans et al. 2023 ). Beyond these broader North Sea–related patterns, the genetic data also reveals two outliers. Males CL032 and CL043 clustered close to modern-day Greek and Estonian populations respectively. Ancestry estimates from McColl et al. (McColl et al. 2024 ) similarly indicated that these two individuals were outliers relative to the major North Sea ancestral component, consistent with our PCA results. CL032 did not have any enamel suitable for isotopic analysis, but isotope analysis was possible for adolescent male CL043. The Sr-O isotope data, however, do not show explicit evidence for mobility during his first 8 years of life. Genetic ancestry does not equate to geographic origin; however, when combined with isotopic evidence, it provides an independent line of evidence for addressing questions of provenance, particularly in the case of such a pronounced outlier. The strontium isotope ratios measured in CL043 (0.71006 and 0.71000 for the first and second molars respectively) are not found in the immediate region of Valkenburg, and the δ ¹⁸O PDB data are too low for the Netherlands (Kootker, Van der Velde, and Heeren 2022 ; Kootker and Heeren 2022) and point toward a childhood spent within a large continental landmass, such as Estonia (–7.8‰ and − 7.9‰ respectively). Interestingly, the age-at-death was determined to be between 13–16 years old (Lonnée and Maat 1998 ), which means that the boy moved to Valkenburg between 8 and 16 years old. Based on a recently published strontium isotope baseline map for Estonia (Suni et al. 2025 ), two possible regions of origins, i.e., coastal and central Estonia (0.7094–0.7147) can be identified. Considering the low δ ¹⁸O PDB as well, central Estonia would be the most likely region. Taken together, the genetic and isotopic evidence support the plausibility of a childhood in the Estonian region. Assessing whether large-distance mobility is specific to inhumed individuals or characteristic of the total cemetery community requires comparison with the isotopic evidence from cremated remains. The comparison between inhumation and cremation isotope data is not straightforward, as the two tissues record different life stages: childhood in enamel versus later life in calcined bone (see (De Coster et al. 2024 ) and references therein). In addition, the contrast between cremated bone, which reflects isotopic signals integrated over periods of skeletal turnover prior to death, and dental enamel from inhumations, which records incremental mineralisation during childhood, result in fundamentally different averaging or mixing intervals. However, with consideration of these crucial differences between the two data sets, important insights are also gained when studying the 87 Sr/ 86 Sr of both the inhumations and cremations at Valkenburg. Mirroring the inhumation data, the cremations exhibit significant variability and includes isotopic ratios that are found across Europe. This indicates that the difference in the social groups using inhumation rather than cremation, cannot be the sole explanation, since both groups appear to consist of mobile individuals. Rather than reflecting fixed distinctions in origin, the coexistence of inhumation and cremation at Valkenburg may point to a funerary landscape shaped by choice, circumstance, and mobility. From the isotopic perspective presented in this study, the distinction between cremation and inhumation has often been treated as a primary structuring principle within burial analysis, reflecting the fact that cremation constituted the dominant funerary norm in much of the Roman world, including the Netherlands and its surrounding regions (Capuzzo et al. 2020). Against this background, inhumation burials in Valkenburg has frequently been approached as an exceptional practice and interpreted as potentially socially or culturally ‘different’. However, this emphasis contrasts with the way other forms of variability within burial practices are generally approached. Considerable variation exists within both cremation and inhumation burials, including differences in grave form, body position, demography, spatial distribution, chronology, and the presence or absence of grave goods. These forms of variability are rarely assigned the same categorical weight as the cremation–inhumation distinction, but are instead understood as reflecting a range of choices made in the treatment of the deceased. From this perspective, the distinction between cremation and inhumation may not be fundamentally different from other observed variations, such as prone versus supine burial or bustum versus urn cremation. All of these practices represent decisions made by the community for a deceased individual, the motivations of which, whether social, cultural, practical, or symbolic, are not always archaeologically accessible. While in certain Roman contexts deviations from the cremation norm may indeed carry interpretative weight (Kootker and Heeren 2022), this may not be necessarily the case in military settings like Valkenburg. In the context of a Roman castellum, cremation and inhumation may have functioned as parallel and socially accepted funerary practices rather than as markers of cultural or social distinction. The explicit separation of cremations and inhumations into distinct analytical groups may therefore overstate the significance of this single aspect of funerary practice, particularly in the absence of clear isotopic, demographic, or contextual evidence supporting such a division, as is the case here. The results presented in this paper suggest that greater caution is warranted in treating inhumation as inherently ‘other’, and that Roman (military) cemeteries may require an analytical approach that places less interpretive weight on funerary rite alone. An alternative interpretation is that the inhumed individuals represent a structurally lower socio-economic segment of the Valkenburg community (Lonnée and Maat 1998 ). The outdated idea of non-local inhumation versus local cremation has been disproven both in this study and in other work (Osterholtz 2025 ; Barberan et al. 2007 ; Nugent 2017 ). Other differences could point toward lower status for the inhumations. Almost all skeletons show indicators of heavy physical labour (Lonnée and Maat 1998 ). Taking into account that these indicators would be less visible in cremations, it remains unsure that this was a specific attribute of the inhumations. Most were buried without grave goods (Smits 2006 ; Lonnée and Maat 1998 ). Importantly, the inverse interpretation (‘inhumation as an indicator of elevated social status’) is not supported at Valkenburg, as no richly furnished inhumation graves have been identified at the site. This study also found no evidence of genealogical relationships indicative of kinship among the inhumed individuals, nor between these individuals and more than 600 others sampled across Europe, supporting the interpretation of a burial scenario lacking interpersonal genetic ties. Although it cannot be established with certainty that such factors directly influenced the community’s decision-making, cremation is generally more time-consuming and sometimes more costly than inhumation (McKinley 2006 ), which may have contributed to the choice of burial for these individuals. The radiocarbon dates (McColl et al. 2024 ) indicate that they died across different periods and therefore do not represent a single contemporaneous episode or exceptional event, nor is it indicative of a family’s specific preferences since no kinship was found. Instead, their treatment appears to reflect a recurring mortuary practice directed toward a particular segment of the population. When considered together with their demographic profile, minimal grave goods, signs of physical strain, and the cost differential between cremation and inhumation, it is also plausible that these inhumed individuals occupied a lower-status position within the broader Valkenburg community. These considerations have important implications for how the biomolecular evidence from Valkenburg should be interpreted. Because the majority of the population was cremated, only a small subset of individuals, namely those who received inhumation, preserved DNA and combined Sr-O-C isotope analyses. Consequently, the genomic results and much of the multi-isotope enamel data may reflect either the life histories of individuals occupying a lower socio-economic position within the community, or a subset of the population differentiated by funerary practice rather than by social or biological distinction alone. Moreover, it should be noted that the available cremation dataset itself represents only a limited subset of the total number of cremation graves, further constraining the representativeness of the biomolecular sample. In contrast, the cremated majority can contribute only strontium isotope ratios, which provide information on childhood mobility but not on ancestry or dietary patterns. The strong variability in strontium signatures observed in both groups nonetheless demonstrate that long distance mobility was a widespread phenomenon within the Valkenburg community. However, the finer-grained genetic and isotopic evidence relating to ancestry, diet, and life-stage mobility reflect the experiences of individuals whose mortuary treatment already sets them apart archaeologically. When interpreting the combined biomolecular dataset, it is therefore important to recognize that the inhumed individuals might not constitute a representative cross-section of the Valkenburg population. Rather, they provide insight into one particular segment of a broader and more heterogeneous borderland population. At the same time, they do prove that some level of long-distance connectivity existed and shaped the community at Valkenburg, indicating that individuals with diverse geographic origins were incorporated into the settlement across multiple generations. Combined, the biomolecular results provide direct evidence for the demographic diversity and mobility that scholars have increasingly recognized in Roman borderland communities (McColl et al. 2024 ; Blöcher et al. 2025 ; Greene 2019 ). Rather than a closed military enclave, Valkenburg appears to have been a place where individuals of different geographic origins, cultural practices and social identities lived in proximity and participated in a shared borderland environment. This picture aligns with recent archaeological work emphasizing the socially mixed character of military settlements, where soldiers, families, labourers, and other non-military individuals formed interdependent communities rather than rigidly separated groups (Allison 2013 ; Walas 2015 ; Greene 2019 ; James 1999 ; Haynes 1999a ). The Valkenburg assemblage offers a tangible biomolecular example of the heterogeneity long suggested in archaeological and historical studies (Haynes 2013 ; Haynes 1999b ), underscoring how borderland populations were fundamentally shaped by movement, interaction, and continual demographic renewal. Conclusion The biomolecular evidence from Valkenburg demonstrates that this Roman borderland settlement became a demographic melting pot, shaped by sustained mobility and population heterogeneity. Integrated isotopic and genetic analyses reveal that individuals of diverse geographic origins were incorporated into a shared community along the Lower Rhine Limes, consistent with emerging views of Roman military settlements as socially entangled rather than homogeneous entities. Strontium and oxygen isotope data indicate that mobility were widespread and occurred across different life stages. Comparable levels of strontium isotope variability in both cremated and inhumed individuals suggest that mobility was not restricted to a particular burial rite. Instead, the coexistence of cremation and inhumation reflects a broader spectrum of funerary choices rather than fixed distinctions in geographic origin. Treating burial rite as a primary analytical division risks overstating its interpretive significance, particularly when other forms of mortuary variability are not assigned equivalent weight. At the same time, the inhumed individuals likely represent a specific segment within this melting pot rather than a separate population. Indicators of heavy physical labour, minimal grave goods, lack of kinship ties, and the relative cost of burial practices point toward social differentiation within the community. While preservation biases limit the representativeness of the biomolecular sample, the strong isotopic variability observed across burial rites nonetheless demonstrates that mobility was a defining feature of the Valkenburg population. Overall, Valkenburg emerges as a borderland community fundamentally shaped by movement, interaction, and continual demographic renewal, providing a biomolecular case study of the diverse and dynamic character of life in the borderlands of the Roman Empire. Declarations Acknowledgements The authors thank Suzan Verdegaal-Warmerdam for conducting the O-C isotope analyses. We are also indebted C-Limes minor Saxion archaeology students for their assistance in sampling the cremated human remains and the creation of an extensive database of the Valkenburg cemetery. Funding statement This study was supported by ‘Constructing the Limes: Employing citizen science to understand borders and border systems from the Roman period until today’ (C-Limes), funded by the Dutch Research Council (NWO) as part of the Dutch Research Agenda (NWA, 2021-2026, project number: NWA.1292.19.364). CrEdiT authorship contribution statement MDC - Conceptualization, Methodology, Validation, Formal analysis, Investigation, Data curation, Writing – original draft. FF - Formal analysis, Writing – original draft. WV – Resource, Writing – editing and reviewing. JO - Writing – editing and reviewing. GRD - Writing – editing and reviewing. LK - Conceptualization, Methodology, Data curation, Funding acquisition, Project administration Writing – editing and reviewing. Data availability All isotope ratio and elemental concentration data generated and analysed during this study are available in the IsoArcH database. References Allison PM (2013) People and Spaces in Roman Military Bases. Cambridge University Press, Cambridge Antonio ML, Gao Z, Moots HM, Lucci M, Candilio F, Sawyer S, Oberreiter V, Calderon D, Devitofranceschi K, Aikens RC, Aneli S, Bartoli F, Bedini A, Cheronet O, Cotter DJ, Fernandes DM (2019) Gabriella Gasperetti, Renata Grifoni, Alessandro Guidi, Francesco La Pastina, Ersilia Loreti, Daniele Manacorda, Giuseppe Matullo, Simona Morretta, Alessia Nava, Vincenzo Fiocchi Nicolai, Federico Nomi, Carlo Pavolini, Massimo Pentiricci, Philippe Pergola, Marina Piranomonte, Ryan Schmidt, Giandomenico Spinola, Alessandra Sperduti, Mauro Rubini, Luca Bondioli, Alfredo Coppa, Ron Pinhasi, and Jonathan K. Pritchard. 'Ancient Rome: A genetic crossroads of Europe and the Mediterranean', Science , 366: 708 – 14 Arcini C (2009) The Worldwide Phenomenon of Ancient Prone Burial.' in Gustin I Back. In: Danielsson IM, Larsson A, Myrberg N, Thedeén S (eds) Döda personers sällskap Gravmaterialens identiteter och kulturellauttryck / On the Threshold Burial Archaeology in the Twenty-First Century. Stockholm University: Stockholm) Auton, Adam, Gonçalo R. Abecasis, David M. Altshuler, Richard M. Durbin, Gonçalo R.Abecasis, David R. Bentley, Aravinda Chakravarti, Andrew G. Clark, Peter Donnelly,Evan E. Eichler, Paul Flicek, Stacey B. Gabriel, Richard A. Gibbs, Eric D. Green,Matthew E. Hurles, Bartha M. Knoppers, Jan O. Korbel, Eric S. Lander, Charles Lee,Hans Lehrach, Elaine R. Mardis, Gabor T. Marth, Gil A. McVean, Deborah A. Nickerson,Jeanette P. Schmidt, Stephen T. Sherry, Jun Wang, Richard K. Wilson, Richard A. Gibbs,Eric Boerwinkle, Harsha Doddapaneni, Yi Han, Viktoriya Korchina, Christie Kovar, Sandra Lee, Donna Muzny, Jeffrey G. Reid, Yiming Zhu, Jun Wang, Yuqi Chang, Qiang Feng, Xiaodong Fang, Xiaosen Guo, Min Jian, Hui Jiang, Xin Jin, Tianming Lan, Guoqing Li, Jingxiang Li, Yingrui Li, Shengmao Liu, Xiao Liu, Yao Lu, Xuedi Ma, Meifang Tang, Bo Wang, Guangbiao Wang, Honglong Wu, Renhua Wu, Xun Xu, Ye Yin, Dandan Zhang, Wenwei Zhang, Jiao Zhao,Meiru Zhao, Xiaole Zheng, Eric S. Lander, David M. Altshuler, Stacey B. Gabriel, Namrata Gupta, Neda Gharani, Lorraine H. Toji, Norman P. Gerry, Alissa M. Resch, Paul Flicek,Jonathan Barker, Laura Clarke, Laurent Gil, Sarah E. Hunt, Gavin Kelman, Eugene Kulesha,Rasko Leinonen, William M. McLaren, Rajesh Radhakrishnan, Asier Roa, Dmitriy Smirnov,Richard E. Smith, Ian Streeter, Anja Thormann, Iliana Toneva, Brendan Vaughan, Xiangqun Zheng-Bradley, David R. Bentley, Russell Grocock, Sean Humphray, Terena James, Zoya Kingsbury, Hans Lehrach, Ralf Sudbrak, Marcus W. Albrecht, Vyacheslav S. Amstislavskiy,Tatiana A. Borodina, Matthias Lienhard, Florian Mertes, Marc Sultan, Bernd Timmermann,Marie-Laure Yaspo, Elaine R. Mardis, Richard K. Wilson, Lucinda Fulton, Robert Fulton,Stephen T. Sherry, Victor Ananiev, Zinaida Belaia, Dimitriy Beloslyudtsev, Nathan Bouk, Chao Chen, Deanna Church, Robert Cohen, Charles Cook, John Garner, Timothy Hefferon,Mikhail Kimelman, Chunlei Liu, John Lopez, Peter Meric, Chris O’Sullivan, Yuri Ostapchuk,Lon Phan, Sergiy Ponomarov, Valerie Schneider, Eugene Shekhtman, Karl Sirotkin, Douglas Slotta, Hua Zhang, Gil A. McVean, Richard M. Durbin, Senduran Balasubramaniam, John Burton, Petr Danecek, Thomas M. Keane, Anja Kolb-Kokocinski, Shane McCarthy, James Stalker, Michael Quail, Jeanette P. Schmidt, Christopher J. Davies, Jeremy Gollub,Teresa Webster, Brant Wong, Yiping Zhan, Adam Auton, Christopher L. Campbell, Yu Kong,Anthony Marcketta, Richard A. Gibbs, Fuli Yu, Lilian Antunes, Matthew Bainbridge,Donna Muzny, Aniko Sabo, Zhuoyi Huang, Jun Wang, Lachlan J. M. Coin, Lin Fang, Xiaosen Guo, Xin Jin, Guoqing Li, Qibin Li, Yingrui Li, Zhenyu Li, Haoxiang Lin, Binghang Liu, Ruibang Luo, Haojing Shao, Yinlong Xie, Chen Ye, Chang Yu, Fan Zhang, Hancheng Zheng, Hongmei Zhu, Can Alkan, Elif Dal, Fatma Kahveci, Gabor T. Marth, Erik P. Garrison,Deniz Kural, Wan-Ping Lee, Wen Fung Leong, Michael Stromberg, Alistair N. Ward, Jiantao Wu, Mengyao Zhang, Mark J. Daly, Mark A. Depristo, Robert E. Handsaker, David M. Altshuler,Eric Banks, Gaurav Bhatia, Guillermo Del Angel, Stacey B. Gabriel, Giulio Genovese,Namrata Gupta, Heng Li, Seva Kashin, Eric S. Lander, Steven A. McCarroll, James C.Nemesh, Ryan E. Poplin, Seungtai C. Yoon, Jayon Lihm, Vladimir Makarov, Andrew G.Clark, Srikanth Gottipati, Alon Keinan, Juan L. Rodriguez-Flores, Jan O. Korbel, Tobias Rausch, Markus H. Fritz, Adrian M. Stütz, Paul Flicek, Kathryn Beal, Laura Clarke,Avik Datta, Javier Herrero, William M. McLaren, Graham R. S. Ritchie, Richard E. Smith,Daniel Zerbino, Xiangqun Zheng-Bradley, Pardis C. Sabeti, Ilya Shlyakhter, Stephen F. Schaffner, Joseph Vitti, David N. Cooper, Edward V. Ball, Peter D. Stenson, David R. Bentley, Bret Barnes, Markus Bauer, R. Keira Cheetham, Anthony Cox, Michael Eberle,Sean Humphray, Scott Kahn, Lisa Murray, John Peden, Richard Shaw, Eimear E. Kenny,Mark A. Batzer, Miriam K. Konkel, Jerilyn A. Walker, Daniel G. Macarthur, Monkol Lek,Ralf Sudbrak, Vyacheslav S. Amstislavskiy, Ralf Herwig, Elaine R. Mardis, Li Ding,Daniel C. Koboldt, David Larson, Kai Ye, Simon Gravel, Anand Swaroop, Emily Chew,Tuuli Lappalainen, Yaniv Erlich, Melissa Gymrek, Thomas Frederick Willems, Jared T.Simpson, Mark D. Shriver, Jeffrey A. Rosenfeld, Carlos D. Bustamante, Stephen B. Montgomery,Francisco M. De La Vega, Jake K. Byrnes, Andrew W. Carroll, Marianne K. Degorter,Phil Lacroute, Brian K. Maples, Alicia R. Martin, Andres Moreno-Estrada, Suyash S.Shringarpure, Fouad Zakharia, Eran Halperin, Yael Baran, Charles Lee, Eliza Cerveira,Jaeho Hwang, Ankit Malhotra, Dariusz Plewczynski, Kamen Radew, Mallory Romanovitch,Chengsheng Zhang, Fiona C. L. Hyland, David W. Craig, Alexis Christoforides, Nils Homer, Tyler Izatt, Ahmet A. Kurdoglu, Shripad A. Sinari, Kevin Squire, Stephen T.Sherry, Chunlin Xiao, Jonathan Sebat, Danny Antaki, Madhusudan Gujral, Amina Noor,Kenny Ye, Esteban G. Burchard, Ryan D. Hernandez, Christopher R. Gignoux, David Haussler,Sol J. Katzman, W. James Kent, Bryan Howie, Andres Ruiz-Linares, Emmanouil T. Dermitzakis,Scott E. Devine, Gonçalo R. Abecasis, Hyun Min Kang, Jeffrey M. Kidd, Tom Blackwell,Sean Caron, Wei Chen, Sarah Emery, Lars Fritsche, Christian Fuchsberger, Goo Jun,Bingshan Li, Robert Lyons, Chris Scheller, Carlo Sidore, Shiya Song, Elzbieta Sliwerska,Daniel Taliun, Adrian Tan, Ryan Welch, Mary Kate Wing, Xiaowei Zhan, Philip Awadalla,Alan Hodgkinson, Yun Li, Xinghua Shi, Andrew Quitadamo, Gerton Lunter, Gil A. McVean,Jonathan L. Marchini, Simon Myers, Claire Churchhouse, Olivier Delaneau, Anjali Gupta-Hinch,Warren Kretzschmar, Zamin Iqbal, Iain Mathieson, Androniki Menelaou, Andy Rimmer,Dionysia K. Xifara, Taras K. Oleksyk, Yunxin Fu, Xiaoming Liu, Momiao Xiong, Lynn Jorde, David Witherspoon, Jinchuan Xing, Evan E. Eichler, Brian L. Browning, Sharon R. Browning, Fereydoun Hormozdiari, Peter H. Sudmant, Ekta Khurana, Richard M. Durbin,Matthew E. Hurles, Chris Tyler-Smith, Cornelis A. Albers, Qasim Ayub, Senduran Balasubramaniam,Yuan Chen, Vincenza Colonna, Petr Danecek, Luke Jostins, Thomas M. Keane, Shane McCarthy,Klaudia Walter, Yali Xue, Mark B. Gerstein, Alexej Abyzov, Suganthi Balasubramanian,Jieming Chen, Declan Clarke, Yao Fu, Arif O. Harmanci, Mike Jin, Donghoon Lee, Jeremy Liu, Xinmeng Jasmine Mu, Jing Zhang, Yan Zhang, Yingrui Li, Ruibang Luo, Hongmei Zhu,Can Alkan, Elif Dal, Fatma Kahveci, Gabor T. Marth, Erik P. Garrison, Deniz Kural,Wan-Ping Lee, Alistair N. Ward, Jiantao Wu, Mengyao Zhang, Steven A. McCarroll, Robert E. Handsaker, David M. Altshuler, Eric Banks, Guillermo Del Angel, Giulio Genovese,Chris Hartl, Heng Li, Seva Kashin, James C. Nemesh, Khalid Shakir, Seungtai C. Yoon,Jayon Lihm, Vladimir Makarov, Jeremiah Degenhardt, Jan O. Korbel, Markus H. Fritz,Sascha Meiers, Benjamin Raeder, Tobias Rausch, Adrian M. Stütz, Paul Flicek, Francesco Paolo Casale, Laura Clarke, Richard E. Smith, Oliver Stegle, Xiangqun Zheng-Bradley,David R. Bentley, Bret Barnes, R. Keira Cheetham, Michael Eberle, Sean Humphray, Scott Kahn, Lisa Murray, Richard Shaw, Eric-Wubbo Lameijer, Mark A. Batzer, Miriam K. Konkel,Jerilyn A. Walker, Li Ding, Ira Hall, Kai Ye, Phil Lacroute, Charles Lee, Eliza Cerveira,Ankit Malhotra, Jaeho Hwang, Dariusz Plewczynski, Kamen Radew, Mallory Romanovitch,Chengsheng Zhang, David W. Craig, Nils Homer, Deanna Church, Chunlin Xiao, Jonathan Sebat, Danny Antaki, Vineet Bafna, Jacob Michaelson, Kenny Ye, Scott E. Devine, Eugene J. Gardner, Gonçalo R. Abecasis, Jeffrey M. Kidd, Ryan E. Mills, Gargi Dayama, Sarah Emery, Goo Jun, Xinghua Shi, Andrew Quitadamo, Gerton Lunter, Gil A. McVean, Ken Chen,Xian Fan, Zechen Chong, Tenghui Chen, David Witherspoon, Jinchuan Xing, Evan E. Eichler,Mark J. Chaisson, Fereydoun Hormozdiari, John Huddleston, Maika Malig, Bradley J.Nelson, Peter H. Sudmant, Nicholas F. Parrish, Ekta Khurana, Matthew E. Hurles, Ben Blackburne, Sarah J. Lindsay, Zemin Ning, Klaudia Walter, Yujun Zhang, Mark B. Gerstein,Alexej Abyzov, Jieming Chen, Declan Clarke, Hugo Lam, Xinmeng Jasmine Mu, Cristina Sisu, Jing Zhang, Yan Zhang, Richard A. Gibbs, Fuli Yu, Matthew Bainbridge, Danny Challis, Uday S. Evani, Christie Kovar, James Lu, Donna Muzny, Uma Nagaswamy, Jeffrey G. Reid, Aniko Sabo, Jin Yu, Xiaosen Guo, Wangshen Li, Yingrui Li, Renhua Wu, Gabor T. Marth, Erik P. Garrison, Wen Fung Leong, Alistair N. Ward, Guillermo Del Angel,Mark A. Depristo, Stacey B. Gabriel, Namrata Gupta, Chris Hartl, Ryan E. Poplin, Andrew G. Clark, Juan L. Rodriguez-Flores, Paul Flicek, Laura Clarke, Richard E. Smith, Xiangqun Zheng-Bradley, Daniel G. Macarthur, Elaine R. Mardis, Robert Fulton, Daniel C. Koboldt,Simon Gravel, Carlos D. Bustamante, David W. Craig, Alexis Christoforides, Nils Homer,Tyler Izatt, Stephen T. Sherry, Chunlin Xiao, Emmanouil T. Dermitzakis, Gonçalo R.Abecasis, Hyun Min Kang, Gil A. McVean, Mark B. Gerstein, Suganthi Balasubramanian,Lukas Habegger, Haiyuan Yu, Paul Flicek, Laura Clarke, Fiona Cunningham, Ian Dunham,Daniel Zerbino, Xiangqun Zheng-Bradley, Kasper Lage, Jakob Berg Jespersen, Heiko Horn,Stephen B. Montgomery, Marianne K. Degorter, Ekta Khurana, Chris Tyler-Smith, Yuan Chen, Vincenza Colonna, Yali Xue, Mark B. Gerstein, Suganthi Balasubramanian, Yao Fu, Donghoon Kim, Adam Auton, Anthony Marcketta, Rob Desalle, Apurva Narechania, Melissa A. Wilson Sayres, Erik P. Garrison, Robert E. Handsaker, Seva Kashin, Steven A. McCarroll,Juan L. Rodriguez-Flores, Paul Flicek, Laura Clarke, Xiangqun Zheng-Bradley, Yaniv Erlich, Melissa Gymrek, Thomas Frederick Willems, Carlos D. Bustamante, Fernando L.Mendez, G. David Poznik, Peter A. Underhill, Charles Lee, Eliza Cerveira, Ankit Malhotra,Mallory Romanovitch, Chengsheng Zhang, Gonçalo R. Abecasis, Lachlan Coin, Haojing Shao, David Mittelman, Chris Tyler-Smith, Qasim Ayub, Ruby Banerjee, Maria Cerezo,Yuan Chen, Thomas W. Fitzgerald, Sandra Louzada, Andrea Massaia, Shane McCarthy, Graham R. Ritchie, Yali Xue, Fengtang Yang, Richard A. Gibbs, Christie Kovar, Divya Kalra,Walker Hale, Donna Muzny, Jeffrey G. Reid, Jun Wang, Xu Dan, Xiaosen Guo, Guoqing Li, Yingrui Li, Chen Ye, Xiaole Zheng, David M. Altshuler, Paul Flicek, Laura Clarke,Xiangqun Zheng-Bradley, David R. Bentley, Anthony Cox, Sean Humphray, Scott Kahn,Ralf Sudbrak, Marcus W. Albrecht, Matthias Lienhard, David Larson, David W. Craig,Tyler Izatt, Ahmet A. Kurdoglu, Stephen T. Sherry, Chunlin Xiao, David Haussler, Gonçalo R. Abecasis, Gil A. McVean, Richard M. Durbin, Senduran Balasubramaniam, Thomas M.Keane, Shane McCarthy, James Stalker, Aravinda Chakravarti, Bartha M. Knoppers, Gonçalo R. Abecasis, Kathleen C. Barnes, Christine Beiswanger, Esteban G. Burchard, Carlos D. Bustamante, Hongyu Cai, Hongzhi Cao, Richard M. Durbin, Norman P. Gerry, Neda Gharani,Richard A. Gibbs, Christopher R. Gignoux, Simon Gravel, Brenna Henn, Danielle Jones,Lynn Jorde, Jane S. Kaye, Alon Keinan, Alastair Kent, Angeliki Kerasidou, Yingrui Li, Rasika Mathias, Gil A. McVean, Andres Moreno-Estrada, Pilar N. Ossorio, Michael Parker, Alissa M. Resch, Charles N. Rotimi, Charmaine D. Royal, Karla Sandoval, Yeyang Su, Ralf Sudbrak, Zhongming Tian, Sarah Tishkoff, Lorraine H. Toji, Chris Tyler-Smith,Marc Via, Yuhong Wang, Huanming Yang, Ling Yang, Jiayong Zhu, Walter Bodmer, Gabriel Bedoya, Andres Ruiz-Linares, Zhiming Cai, Yang Gao, Jiayou Chu, Leena Peltonen, Andres Garcia-Montero, Alberto Orfao, Julie Dutil, Juan C. Martinez-Cruzado, Taras K. Oleksyk,Kathleen C. Barnes, Rasika A. Mathias, Anselm Hennis, Harold Watson, Colin McKenzie,Firdausi Qadri, Regina Larocque, Pardis C. Sabeti, Jiayong Zhu, Xiaoyan Deng, Pardis C. Sabeti, Danny Asogun, Onikepe Folarin, Christian Happi, Omonwunmi Omoniwa, Matt Stremlau, Ridhi Tariyal, Muminatou Jallow, Fatoumatta Sisay Joof, Tumani Corrah, Kirk Rockett, Dominic Kwiatkowski, Jaspal Kooner, Trâ N. Tịnh Hiê`N, Sarah J. Dunstan,Nguyen Thuy Hang, Richard Fonnie, Robert Garry, Lansana Kanneh, Lina Moses, Pardis C. Sabeti, John Schieffelin, Donald S. Grant, Carla Gallo, Giovanni Poletti, Danish Saleheen, Asif Rasheed, Lisa D. Brooks, Adam L. Felsenfeld, Jean E. McEwen, Yekaterina Vaydylevich, Eric D. Green, Audrey Duncanson, Michael Dunn, Jeffery A. Schloss, Jun Wang, Huanming Yang, Adam Auton, Lisa D. Brooks, Richard M. Durbin, Erik P. Garrison,Hyun Min Kang, Jan O. Korbel, Jonathan L. Marchini, Shane McCarthy, Gil A. McVean,and Gonçalo R. Abecasis. 2015. 'A global reference for human genetic variation', Nature , 526: 68–74 Barberan Sébastien, Bonnet C, Blaizot F, Wittman A (2007) 'nhumation and cremation in Roman Gaul: continuity or discontinuity of the burial practices', Schriften des Archäologischen Museums Frankfurt , 21: 305 – 22 Blöcher J, Vallini L, Velte M, Eckel R, Guyon Léa, Winkelbach L, Thomas MG, Gharehbaghi N, Mitchell CT, Schümann J, Köhler S, Seyr E, Krichel K, Rau S, Hirsch J, Duras J, Klement K (2025) Miriam Wilkenhöner, Lisa Vetterdietz, Francesca Gentilin, Melany Müller, Anna-Lena Mücke, Nicoletta Zedda, Youssef Tawfik, Eveline Saal, George McGlynn, Barbara Bramanti, Jörg Orschiedt, Regina Molitor, Barbara Fliß, Ines Spazier, David Shankland, Claus Vetterling, Kurt Karpf, Vera Planert, Stefan Hölzl, Silvia Codreanu-Windauer, Dieter Quast, Ilija Mikić, Bernd Päffgen, Maxime Brami, Thomas Richter, Raphaëlle Chaix, Susanne Brather-Walter, Peter Steffens, Markus Marquart, Thomas Becker, Jochen Haberstroh, Mischa Meier, Sebastian Schmidt-Hofner, Sebastian Brather, Michaela Harbeck, Steffen Patzold, Daniel Wegmann, and Joachim Burger. Historic Genomes Uncover Demographic Shifts and Kinship Structures in Post-Roman Central Europe. In.: Cold Spring Harbor Laboratory Breeze DJ, Dobson B (1987) Hadrian's Wall (Penguin: London) Capuzzo G, Snoeck C, Boudin M, Dalle S, Annaert R, Hlad M, Kontopoulos I, Sabaux C, Salesse K, Sengeløv A, Stamataki E, Veselka B Eugène Warmenbol, Guy De Mulder, Dries Tys, and Martine Vercauteren. 2020. 'CREMATION VS. INHUMATION: MODELING CULTURAL CHANGES IN FUNERARY PRACTICES FROM THE MESOLITHIC TO THE MIDDLE AGES IN BELGIUM USING KERNEL DENSITY ANALYSIS ON < sup>14 C DATA'. Radiocarbon, 62: 1809–1832 Cassidy LM, Russell M, Smith M, Delbarre G, Cheetham P, Manley H, Mattiangeli V, Breslin EM Iseult Jackson, Maeve McCann, Harry Little, Ciarán G. O’Connor, Beth Heaslip, Daniel Lawson, Phillip Endicott, and Daniel G. Bradley. 2025. 'Continental influx and pervasive matrilocality in Iron Age Britain'. Nature, 637: 1136–1142 Chaitanya L, Breslin K, Zuñiga S, Wirken L, Pośpiech E, Kukla-Bartoszek M, Sijen T, Knijff PD, Liu F, Branicki W, Kayser M, Susan Walsh (2018) The HIrisPlex-S system for eye, hair and skin colour prediction from DNA: Introduction and forensic developmental validation. Forensic Sci International: Genet 35:123–135 Chenery C, Müldner G, Evans JA, Eckardt H, Lewis M (2010) Strontium and stable isotope evidence for diet and mobility in Roman Gloucester, UK. J Archaeol Sci 37:150–163 Collins R, and Frances McIntosh ^ (ed) (2014) Life in the Limes. Studies of the people and objets of the Roman frontiers presented at LindsayAllason-Jones on the occasion of her birthday and retirement. Oxford, Oxbow Danecek P, Auton A, Abecasis G, Albers CA, Banks E, Depristo MA, Handsaker RE, Lunter G, Marth GT, Sherry ST Gilean McVean, and Richard Durbin. 2011. 'The variant call format and VCFtools'. Bioinformatics, 27: 2156–2158 De Coster M, Ammer T, Laning, Kootker LM (2024) The Relevance of Sr–O–C Isotope Analysis on Burnt Human Skeletal Remains in Archeological and Forensic Contexts: A Review and Future Directions. WIREs Forensic Sci 6:e1524 De Coster MRAL, Kootker LM (2024) 'Chromatographic separation of strontium in archaeological cremated remains for Thermal Ionisation Mass Spectrometry (TIMS) analysis', protocols.io Depaermentier MLC, Krause-Kyora B, Hajdas I, Kempf M, Kuhn T, Spichtig N, Schwarz P-A, and Claudia Gerling (2023). 'Bioarchaeological analyses reveal long-lasting continuity at the periphery of the Late Antique Roman Empire', iScience , 26: 107034 Dierendonck RM, Van DP, Hallewas, Waugh KE (eds) (1993) ^(eds). The Valkenburg Excavations 1985–1988. Introduction and Detail Studies (Amersfoort) Eckardt H, Müldner G, Lewis M (2014) People on the move in Roman Britain. World Archaeol 46:534–550 Evans JA, Mee K, Chenery CA, Marchant AP (2023) Biosphere Isotope Domains GB (V2): Interactive Website. British Geological Survey. (Dataset). In Gardner A (2016) 'An Archaeology of Identity' Greene EM (2019) Roman Military Communities and the Families of Auxiliary Soldiers. In: Brice L (ed) New Approaches to Greek and Roman Warfare. Wiley Haak W, Lazaridis I, Patterson N, Rohland N, Mallick S, Llamas B, Brandt G, Nordenfelt S, Harney E, Stewardson K, Fu Q, Mittnik A, Bánffy E, Economou C, Francken M, Susanne Friederich, Rafael Garrido Pena, Fredrik Hallgren, Valery Khartanovich, Aleksandr Khokhlov, Michael Kunst, Pavel Kuznetsov, Harald Meller, Oleg Mochalov, Vayacheslav Moiseyev, Nicklisch N, Pichler SL, Risch R, Manuel A Rojo Guerra, Christina Roth, Anna Szécsényi-Nagy, Joachim Wahl, Matthias Meyer, Johannes Krause, Dorcas Brown, David Anthony, Alan Cooper, Kurt Werner Alt, and David Reich. 2015. 'Massive migration from the steppe was a source for Indo-European languages in Europe', Nature , 522: 207 – 11 Harvig L, Frei KM, Price TD, Lynnerup N (2014) Strontium Isotope Signals in Cremated Petrous Portions as Indicator for Childhood Origin. PLoS ONE 9:e101603 Haynes I (1999a) Introduction: the Roman army as a community. J Roman Archaeol supplementary Ser 34:7–14 Haynes I (2013) Blood of the Provinces. The Roman Auxilia and the Making of Provincial Society from Augustus to the Severans. Oxford University Press, Oxford Haynes I (1999b) 'Military service and cultural identity - in the auxilia'. J Roman Archaeol, 34 Hoffmann B (2013) The Roman Invasion of Britain. Archaeology versus History , (Barnsley) James S (1999) 'The Community of the Soldiers: A Major Identity and Centre of Power in the Roman Empire', Theoretical Roman Archaeology Journal , 1998: 14–25 Johnson A (1983) Roman forts of the 1st and 2nd centuries AD in Britain and the German provinces. St. Martin's: New York) Kootker LM, De Coster MRAL (2024) 'Chromatographic separation of strontium in archaeological human and faunal enamel for Thermal Ionisation Mass Spectrometry (TIMS) analysis', protocols.io Kootker LM, Van der Velde HM, Heeren S (2022) A triple isotope approach (Sr-O-C) to assess human mobility dynamics in the Lower Germanic limes borderscape (40–470 CE). J Archaeol Science: Rep 44:103520 Kootker L, and Stijn Heeren (2022) Human mobility in the post-Roman Netherlands: combined Sr-O isotopic evidence and archaeological contextualization. Amersfoort, Cultural Heritage Agency Kootker LM, Rowin J, Van Lanen H, Kars, Davies GR (2016) Strontium isoscapes in The Netherlands. Spatial variations in 87 Sr/ 86 Sr as a proxy for palaeomobility. J Archaeol Science: Rep 6:1–13 Lonnée HA, Maat GJR (1998) 'Inhumations in a Roman Cemetery At Valkenburg-Marktveld (Zuid Holland) in the Netherlands'. Barge’s Anthropologica, 3 Madgwick R, Lewis J, Grimes V, Guest P (2019) On the hoof: exploring the supply of animals to the Roman legionary fortress at Caerleon using strontium (87Sr/86Sr) isotope analysis. Archaeol Anthropol Sci 11:223–235 Mallick S, Micco A, Mah M, Ringbauer H, Lazaridis I Iñigo Olalde, Nick Patterson, and David Reich. 2024. 'The Allen Ancient DNA Resource (AADR) a curated compendium of ancient human genomes'. Sci Data, 11 Manley J (2007) AD 43. The Roman Invasion of Britain, a reassessment (Chalford) McColl H, Kroonen G, Víctor Moreno-Mayar J, Seersholm FV, Scorrano G, Pinotti T, Vimala T, Sindbæk SørenM, Ethelberg P, Fyfe R, Gaillard M-J, Hanne M, Ellegård Larsen MF, Mortensen F, Demeter, Marie Louise S, Jørkov S, Bergerbrant PDB, Damgaard ME, Allentoft L, Vinner C, Gaunitz A, Ramsøe I, Altinkaya RA, Henriksen, Evan K, Irving-Pease K, Sjögren S, Sabatini A, Fischer W, Barrie A, Rosengren A, Vaughn J, Cao J, Staring (2024) Jesper Stenderup, Fulya Eylem Yediay, Torbjörn Ahlström, Irakli Akhvlediani, Sofie Laurine Albris, Biyaslan Atabiev, E. V. Balanovskaya, Pernille Bangsgaard, Maria Giovanna Belcastro, Nick Card, Philippe Charlier, Elizaveta Chernykh, Torben Trier Christiansen, Alfredo Coppa, Maura De Coster, Sean Dexter Denham, Sophie Desenne, Jane Downes, Karin Margarita Frei, Olivér Gábor, U. B. Gadiev, Johan Zakarias Gårdsvoll, Zanette Tsigaridas Glørstad, Jesper Hansen, Stijn Heeren, Merete Henriksen, Volker Heyd, Mette Høj, Mads Kähler Holst, Rimantas Jankauskas, Henrik Janson, Mads Dengsø Jessen, Jens Winther Johannsen, Torkel Johansen, Ole Thirup Kastholm, Anton Kern♰, Ruslan Khaskhanov, Katrine Ipsen Kjær, Vladimir Kolosov, Lisette M. Kootker, Klaudia Kyselicová, Anne Christine Larsen, Thierry Lejars, Mette Løvschal, Niels Lynnerup, Yvonne Magnusson, V. Yu Malashev, Kristiina Mannermaa, Vyacheslav Masyakin, Anne Lene Melheim, Inga Merkyte, Vyacheslav Moiseyev, Stig Bergmann Møller, Erika Molnár, Nadja Mortensen, Eileen Murphy, Bjarne Henning Nielsen, Doris Pany-Kucera, Bettina Schulz Paulsson, Gertjan Plets, Marcia S. Ponce De León, Håkon Reiersen, Walter Reinhard, Antti Sajantila, Birgitte Skar, Vladimir Slavchev, Václav Smrčka, Lasse Sørensen, Georg Tiefengraber, Otto Christian Uldum, Helle Vandkilde, Jorge Vega, Daniele Vitali, Alexey Voloshinov, Sidsel Wåhlin, Holger Wendling, Anna Wessman, Helene Wilhelmson, Karin Wiltschke, João Zilhão, Christoph Pe Zollikofer, Thorfinn Sand Korneliussen, Bruno Chaume, Jean-Paul Demoule, Thomas Werge, Line Olsen, Rasmus Nielsen, Lotte Hedeager, Kristian Kristiansen, Martin Sikora, and Eske Willerslev. Steppe Ancestry in Western Eurasia and the Spread of the Germanic Languages. In.: Cold Spring Harbor Laboratory McKinley JI (2006) Cremation… the cheap option. Social Archaeol funerary remains 1:81–88 Milella M, Mariotti V, Belcastro MG, Knüsel CJ (2015) Patterns of Irregular Burials in Western Europe (1st-5th Century A.D). PLoS ONE 10:e0130616 Mols S (2020) and Rien Polak. 'De Romeinse Limes in Nederland', Lampas , 53: 113 – 23 Noy D (2000) Building a Roman Funeral Pyre. Antichthon 34:30–45 Nugent R (2017) Two of a Kind: Conceptual Similarities between Cremation and Inhumation in Early Anglo-Saxon England.' in Jessica Cerezo-Román. In: Wessman A, Williams H (eds) Cremation and the Archaeology of Death. Oxford University Press Olalde Iñigo, Carrión P, Mikić I, Rohland N, Mallick S, Lazaridis I, Mah M, Korać M, Golubović Snežana, Petković S, Miladinović-Radmilović Nataša, Vulović D, Alihodžić T, Ash A, Baeta M, Bartík J, Bedić Željka, Bilić M, Bonsall C, Bunčić M, Bužanić D, Carić M (2023) Lea Čataj, Mirna Cvetko, Ivan Drnić, Anita Dugonjić, Ana Đukić, Ksenija Đukić, Zdeněk Farkaš, Pavol Jelínek, Marija Jovanovic, Iva Kaić, Hrvoje Kalafatić, Marijana Krmpotić, Siniša Krznar, Tino Leleković, Marian M. De Pancorbo, Vinka Matijević, Branka Milošević Zakić, Anna J. Osterholtz, Julianne M. Paige, Dinko Tresić Pavičić, Zrinka Premužić, Petra Rajić Šikanjić, Anita Rapan Papeša, Lujana Paraman, Mirjana Sanader, Ivana Radovanović, Mirjana Roksandic, Alena Šefčáková, Sofia Stefanović, Maria Teschler-Nicola, Domagoj Tončinić, Brina Zagorc, Kim Callan, Francesca Candilio, Olivia Cheronet, Daniel Fernandes, Aisling Kearns, Ann Marie Lawson, Kirsten Mandl, Anna Wagner, Fatma Zalzala, Anna Zettl, Željko Tomanović, Dušan Keckarević, Mario Novak, Kyle Harper, Michael McCormick, Ron Pinhasi, Miodrag Grbić, Carles Lalueza-Fox, and David Reich. 'A genetic history of the Balkans from Roman frontier to Slavic migrations', Cell , 186: 5472-85.e9 Osterholtz A (2025) Inhumation and Cremation.' in. Springer Nature Switzerland Ota L (2007) Busta in Moesia Inferior. Acta Terrae Septemcastrensis 6:75–99 Patterson N, Moorjani P, Luo Y, Mallick S, Rohland N, Zhan Y (2012) Teri Genschoreck, Teresa Webster, and David Reich. Genetics 192:1065–1093 'Ancient Admixture in Human History' Patterson N, Price AL, Reich D (2006) Population Structure and Eigenanalysis. PLoS Genet 2:e190 Peddie J (2005) Conquest. The Roman invasion of Britain (Sparkford) Perretti S, Santos Patrícia, Vizzari MT, Tassani E, Benazzo A, Ghirotto S, and Guido Barbujani (2025). 'Inference of human pigmentation from ancient DNA by genotype likelihoods', Proceedings of the National Academy of Sciences , 122 Petculescu L (2025) Roman Military Equipment in the Northeastern Part of Moesia Inferior. I. The Capidava Cemetery. Cercetări Arheologice 32:183–200 Philips J, and Pete Wilson (2021) Margary RR81a and a Bustum Burial at Brooklyn House, Norton-on-Derwent, North Yorkshire. Itinera J Roman Roads Res Association 1:207–2018 Plomp E, Von Holstein ICC, Kootker LM, Suzanne JA, Verdegaal-Warmerdam T, Forouzanfar, Davies GR (2020) Strontium, oxygen, and carbon isotope variation in modern human dental enamel. Am J Phys Anthropol 172:586–604 Polak M, Kooistra LI (2013) 'A sustainable frontier? The establishment of the Roman frontier in the Rhine delta', Jahrbuch des Römisch-Germanischen Zentralmuseums , 60 Price AL, Nick J, Patterson RM, Plenge ME, Weinblatt NA, Shadick, Reich D (2006) Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet 38:904–909 Ringbauer H, Huang Y, Akbari A, Mallick S, Olalde Iñigo, Patterson N, David Reich (2024) Accurate detection of identity-by-descent segments in human ancient DNA. Nat Genet 56:143–151 Rubinacci S, Hofmeister RJ (2023) Imputation of low-coverage sequencing data from 150,119 UK Biobank genomes. Nat Genet 55:1088–1090Bárbara Sousa Da Mota, and Olivier Delaneau Scheib CL, Hui R, Rose AK, Inskip ED’AtanasioSA, Dittmar J, Cessford C, Griffith SJ, Solnik A, Wiseman R, Neil B, Biers T, Harknett S-J, Sasso S, Biagini SA, O’Connell C Göran Runfeldt, Corinne Duhig, Christopher Evans, Mait Metspalu, Martin J. Millett, Tamsin, John E. Robb, and Toomas Kivisild. 2024. 'Low Genetic Impact of the Roman Occupation of Britain in Rural Communities'. Mol Biol Evol, 41 Sengeløv A, Capuzzo G, Dalle S, James HF, Sabaux C, Stamataki E, Hlad M, Carina T, Gerritzen EM, Legrand B, Veselka G, De Mulder (2025) Rica Annaert, Mathieu Boudin, Kevin Salesse, Eugène Warmenbol, Nadine Mattielli, Christophe Snoeck, and Martine Vercauteren. 'From plants to patterns: Constructing a comprehensive online strontium isoscape for Belgium (IsoBel) using high density grid mapping', Geoderma , 453: 117123 Shaw H, Montgomery J, Redfern R, Gowland R, Evans JA (2016) Identifying migrants in Roman London using lead and strontium stable isotopes. J Archaeol Sci 66:57–68 Siva N (2008) 1000 Genomes project. Nat Biotechnol 26:256–256 Smits L (2006) Leven en Sterven langs de Limes. Het fysisch-antropologisch onderzoek van vier grafveldpopulaties uit de noordelijke grenszone van Germania inferior in de Vroeg- en Midden-Romeinse tijd. Universiteit van Amsterdam Snoeck C, Lee-Thorp J, Schulting R, De Jong J, Debouge W, Mattielli N (2015) Calcined bone provides a reliable substrate for strontium isotope ratios as shown by an enrichment experiment. Rapid Commun Mass Spectrom 29:107–114 Snoeck C, Pouncett J, Claeys P, Goderis S, Mattielli N, Pearson MP, Willis C, Zazzo A, Lee-Thorp JA, Schulting RJ, Snoeck C, Pouncett J, Claeys P, Goderis S Nadine Mattielli, Mike Parker Pearson, Christie Willis, Antoine Zazzo, Julia A. Lee-Thorp, and Rick J. Schulting. 2018. 'Strontium isotope analysis on cremated human remains from Stonehenge support links with west Wales'. Sci Rep, 8: 10790 Suni R, Kirsimäe K, Rannamäe E, Lõugas L, Maldre L, Tõrv M, Kriiska A, Ester Oras (2025) Defining the Estonian strontium isoscape for archeological and paleoecological provenance studies. Est J Archaeol 29:109–137 Taylor A (2008) Aspects of Deviant Burial in Roman Britain. In: Murphy EM (ed) Deviant Burial in the archaeological Record. Oxford, Oxbow Books Verhagen P, Joyce J, Groenhuijzen MR (2019) Finding the Limits of the Limes: Setting the Scene.' in. Springer International Publishing) Vos W, Blom E, Cornelisse B, de Bruin J, Loopik J, Ruiter A, Tol A and Lourens Van der Feijst. 2021. 'A legionary fortress at Valkenburg: discovering a springboard for the conquest of Britannia?', World Archaeology Vos WK, and Jeroen Oosterbaan. in prep. 'Valkenburg Marktveld (1985–1988) revisited: new insights after the reorganizationreassessment of the largest cemetery along the Dutch part of the Lower German Limes' Walas AH (2015) Roman Military Bases as Social Spaces. University of Leicester Walsh S, Liu F, Wollstein A, Kovatsi L, Ralf A, Kosiniak-Kamysz A, Branicki W, Kayser M (2013) The HIrisPlex system for simultaneous prediction of hair and eye colour from DNA. Forensic Sci International: Genet 7:98–115 Willmes M, Bataille CP, Hannah F, James I, Moffat L, McMorrow L, Kinsley RA, Armstrong Stephen Eggins, and Rainer Grün. 2018. 'Mapping of bioavailable strontium isotope ratios in France for archaeological provenance studies'. Appl Geochem, 90: 75–86 Wynia SL (1999) 'Caius was here. The Emperor Caius’ Preparations for the Invasion of Britannia: New Epigraphic Evidence.' in H. Sarfatij, W.J.H. Verwers and P.J. Woltering (eds.), Discussion with the Past. Archaeological studies presented to W.A. van Es (Amersfoort) Zhao H, Wang ZSJ, Huang H, Kocher J-P, Wang L (2014) CrossMap: a versatile tool for coordinate conversion between genome assemblies. Bioinformatics 30:1006–1007 Additional Declarations No competing interests reported. Supplementary Files ValkenburgSupplementaryTableAASciences.xlsx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 08 Feb, 2026 Editor assigned by journal 27 Jan, 2026 Submission checks completed at journal 26 Jan, 2026 First submitted to journal 26 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information

Related documents

• Fall in Love : Roman | Seasonal it is - die perfekte Lektüre für die ersten Herbststürme! / Romana Stauffer#619630
• Nora Sim - Der Scheriff von Dreistadt Teil I und II : Western Roman / Karina Nicole Brauer#487517
• Data publication: Geochemistry and fluid inclusion analyses of the Sadisdorf Li-Sn-(W-Cu) magmatic-hydrothermal greisen and vein system, eastern Erzgebirge, Germany#647834
• Helium exploration in European rift systems: Regional framework and new data from the Upper Rhine Graben#408761
• The Heart of Rome: A Tale of the "Lost Water"#33535