Primary glass production, the industrial-scale melting of raw materials into slabs of unworked glass that were later shipped to vessel-making workshops across the Mediterranean, was one of the great chemical enterprises of antiquity. During the first millennium CE, it was concentrated almost exclusively in the eastern Mediterranean, where workshops combined quartz-rich sand with natron, a sodium carbonate evaporite harvested almost entirely from Egypt’s Wadi Natrun, to produce the famous Roman glass that filled the empire’s cups, windows, and lamps. The industry operated according to highly consistent recipes, delivering glass of remarkably uniform quality over decades and even centuries, a feat that scholars regard as one of the earliest achievements of organized chemical and industrial engineering. Yet because furnaces were systematically dismantled after each production cycle and materials were consumed with great efficiency, clear archaeological evidence of primary production is extremely rare, and only a handful of sites have ever been identified along the eastern Mediterranean coast, in Israel and Egypt.

The Yavne discovery, made within a vast multi-period industrial landscape that also contained winepresses, pottery kilns, and tombs, is exceptional on several counts. In an area designated Area L, excavators revealed a structure they have dubbed the “Glass House,” measuring at least five by eight meters, which uniquely contained furnaces from both stages of glass production side by side: two large rectangular tank furnaces, each roughly two by four meters, used for melting raw glass, and four small circular furnaces for manufacturing vessels. It is the first in situ complex in the region where primary and secondary glass furnaces were found together within a single building. The furnace floors, built from hard carbonate material, preserved embedded glass lumps, and beneath them lay thick layers of dense black ash, testimony to the enormous quantities of fuel that fed the melt. Excavators believe the furnaces were deliberately dismantled after their working life, which is precisely why so few such installations survive anywhere in the ancient world.

The research team faced a rare opportunity. The site sits within about ten kilometers of five distinct quartz-rich sandy deposits, including modern beach sand, the semi-stabilized dunes of the Palmachim field, cemented aeolianite sandstone known locally as kurkar, buried fluvial sand beneath agricultural fields near the furnaces, and the clay-rich soils of the Soreq stream floodplain. All of these sediments ultimately derive from the Nile River, whose sands have been swept eastward along the Levantine coast by longshore drift since the Pliocene, so they share a common geological ancestry. That geographic accident allowed the scientists to do something few studies have attempted: a direct, side-by-side comparison of finished glass from a primary furnace with the actual raw materials potentially available to the glassmakers who worked there.

The analytical program was exhaustive. Twelve glass samples of various color shades were collected from the furnace tank floor, a fill south of the furnace, and an adjacent stone collapse, along with four slag specimens consisting of vitrified furnace material. Sixteen sediment samples were taken from the five deposit types along a west-to-east transect. Glass and sediment were digested in clean laboratories using multi-acid total dissolution and measured by inductively coupled plasma optical emission spectrometry and mass spectrometry, calibrated against certified reference materials, with procedural blanks held below one percent of measured values. Mineralogy was determined by X-ray diffraction, grain-size distributions by laser diffraction, and, in a methodological first for natron glass studies, stable carbon isotopes of organic carbon trapped within the glass matrix were measured by elemental analysis coupled to isotope-ratio mass spectrometry. Statistical treatment followed best practice for compositional data, with centered log-ratio transformation and z-score normalization feeding into principal component analyses of major elements, trace elements, and rare earth elements treated as separate datasets.

The chemistry told a nuanced story. The Yavne glass displays the overall hallmarks of Levantine natron glass, with sodium oxide content between roughly 11.6 and 16.3 percent, silica estimated between 68.5 and 76.1 percent, and chondrite-normalized rare earth element patterns nearly indistinguishable from the surrounding sands, confirming a shared Nilotic provenance. Yet the Yavne assemblage forms its own distinct compositional group, which the authors propose to call the Levantine-Yavne type. Its characteristic fingerprint, approximately 13.5 percent sodium oxide, 9.5 percent calcium oxide, 3 percent alumina, 0.1 percent titanium, and 0.5 percent iron, resembles the glass of the well-known fourth-century workshop at Jalame but is measurably different in its sodium, titanium, iron, and zirconium contents, and it plots clearly apart from Egyptian glass groups in ratio diagrams of calcium oxide versus alumina and titanium versus alumina. The soda levels are also lower than in earlier Roman glasses, hinting that Yavne stands at the beginning of a compositional trend that developed in later centuries.

The central puzzle emerged when the glass was compared with the candidate sands. The sodium and silica contents of the glass could, in principle, be achieved by melting natron with any of the local deposits, and the best overall chemical match is the orange-tinted dune sand of the Palmachim field several kilometers to the west, which agrees with the glass in its magnesia, lime, and alkali contents. But the Yavne glass is anomalously enriched in aluminum, potassium, and, to a lesser degree, phosphorus and rare earths, and no studied sand deposit, and not even the famously alumina-rich sand of the Belus River further north, can account for that signature on its own. Principal component analysis of trace elements reveals a striking triangular geometry, with the sandy deposits, the glass, and the Soreq floodplain soils occupying the three apexes, and the furnace slags plotting between them. The authors weigh two explanations. Either the glassmakers deliberately added a previously unreported ingredient, most plausibly the clay-rich alluvial soils of the Soreq floodplain, which contain quartz silt and abundant clays rich in aluminum and potassium, or the excess aluminum and potassium reflect contamination of the analyzed glass chunks by furnace material. A third possibility is that an as-yet unsampled sand source was used. Whichever proves correct, the study is the first to raise the possibility that natron glass, long believed to be a simple two-ingredient recipe of quartz sand and natron, may sometimes have contained a third component.

The stable isotope work added an entirely new dimension to the investigation. Organic carbon trapped in the glass ranges from 0.02 to 0.07 percent by weight, with delta carbon-13 values between minus 26.7 and minus 22.6 per mil relative to the Vienna Pee Dee Belemnite standard, implying that somewhere between about 70 and 100 percent of the fuel carbon derived from C3-type vegetation, the trees and shrubs whose photosynthetic pathway yields characteristic isotopically light carbon. Slag samples, richer in organic carbon at 0.38 to 0.56 percent, show even lighter and tightly clustered values around minus 28 per mil, indicating an essentially pure C3 fuel source. The researchers suggest that debris from Yavne’s extensive wine and olive presses, abundant across the industrial complex, may have fed the furnaces. Notably, the mean carbon isotope value of the glass, about minus 24.7 per mil, differs significantly from that of the slag, about minus 28 per mil, hinting at a more complex carbon incorporation process during melting that the authors say deserves further study.

The isotopic data even offered a crude thermometer. Because calcite decomposes to quicklime above about 700 degrees Celsius, and re-carbonated lime inherits strong kinetic isotope fractionation that grows more negative with firing temperature, the carbon isotope values of carbonate in the slag can serve as an “isotopic thermometer.” Extrapolating an established calibration from fired pottery clays, the most negative slag value of minus 23 per mil corresponds to an equivalent temperature of roughly 550 to 800 degrees Celsius. Since natron glass melts require higher temperatures, the authors interpret these figures as minimum estimates and conclude that the sampled slag likely came from cooler, peripheral parts of the furnace rather than the hottest melting zones.

Beyond its technical findings, the study carries broader implications for how archaeologists reconstruct ancient supply chains. What was once imagined as a single sand quarry behind each compositional glass group is now understood as a mosaic of geographically variable deposits and production sites, and the Yavne results demonstrate that meaningful answers require sampling not just the glass and the obvious beach sands but also dunes, buried sediments, kurkar, and soils. The southernmost known primary production site in Israel, Yavne now defines a new compositional group in the Levantine glass landscape and, perhaps more provocatively, opens the possibility that Roman glassmakers were more experimental chemists than their reputation for rigid standardization suggests. As the authors note, only additional sampling of sedimentary deposits around other primary production sites will reveal whether the Yavne recipe was a local innovation or a lost chapter of a much older tradition.

The new study, published in the open-access journal PLOS One on September 9, 2026, was conducted by David Varricchio of Montana State University and Heath Caldwell of North Carolina State University. Together, they examined a remarkably complete growth series of pelvic and hindlimb fossils drawn from animals that lived roughly 75 to 78 million years ago, during the late Cretaceous period, when the region they inhabited was a lush coastal plain bordering an inland seaway.

Troodontids were among the most distinctive dinosaurs of their era: small to medium-sized predators with long, slender legs, unusually large brains and eyes for reptiles, and bodies covered in feathers. Their closest relatives include the dromaeosaurids—the famed “raptors”—and their lineage sits tantalizingly close to the origin of birds. Yet despite their scientific importance, troodontids have long remained shadowy figures in the North American fossil record, known primarily from isolated teeth, fragmentary bones, and the occasional egg clutch.

That scarcity of material has created a persistent problem. The genus Troodon was established in 1856 on the basis of a single tooth, which has served as the type specimen—the anatomical benchmark against which all other members of the species must be judged—ever since. Teeth are notoriously poor anchors for species-level classification among closely related theropod dinosaurs, whose dentition can be remarkably similar across different genera. As a result, the very validity of the Troodon genus has been questioned repeatedly over the decades, with some researchers arguing the name should be abandoned entirely.

The Montana fossils change that calculus dramatically. The material analyzed by Varricchio and Caldwell comes from the upper member of the Two Medicine Formation, a geologically rich sedimentary sequence that has previously yielded eggs, nests, embryos, and adult skeletal remains attributed to a single troodontid population. Because the assemblage spans the full ontogenetic range—from embryonic individuals still inside their eggs to skeletally mature adults—the researchers could, for the first time, reconstruct how the species’ hip and leg bones changed shape, texture, and proportion throughout an individual’s life.

Among the most informative patterns documented in the study is a systematic shift in limb proportions. The ratio of femur length to tibia length, the researchers found, was consistently higher in adult specimens than in juveniles. In practical terms, this means that as these animals matured, their lower leg bones grew proportionally longer relative to their thigh bones—a developmental trajectory familiar from many fast-running animals, both extinct and modern. Such a shift suggests that young troodontids may have moved differently from adults, potentially relying on different locomotor strategies while their bodies were still developing.

Equally significant was the observation that many of these developmental changes occurred at consistent growth stages across multiple individuals. That consistency implies a strong genetic or intrinsic developmental program, relatively insulated from environmental noise. In paleontology, where distinguishing true species differences from individual or environmental variation is a constant challenge, this kind of reproducibility is gold: it demonstrates that the observed anatomical features are stable, species-level characteristics rather than artifacts of habitat, diet, or climate fluctuation.

The study also confirmed two other biological insights. First, adults of the species varied distinctly in body size, hinting at a level of individual variation—or perhaps sexual dimorphism—that future researchers will need to account for when assessing fragmentary troodontid remains. Second, in line with earlier research, the analysis suggests that foot injuries and pathologies may have been remarkably common in the species. Whether these stemmed from high-activity lifestyles, aggressive behavior, nesting-related stress, or something else entirely remains an open question, but the recurrence of foot problems across individuals adds an intriguing dimension to reconstructions of troodontid daily life.

Beyond the biological findings, the study carries a significant taxonomic proposal. Varricchio and Caldwell argue that a specific subset of the remains—a collection of pelvic and hindlimb fossils representing both juveniles and adults—should be formally designated as the type specimen of T. formosus, displacing the original single tooth that has anchored the species since the nineteenth century. Such a move would give the species a robust, diagnosable skeletal foundation and could go a long way toward securing the validity of the Troodon genus itself. However, the authors caution that this reassignment is not yet settled. The proposal depends on a decision currently pending before the International Commission on Zoological Nomenclature, the body that adjudicates disputes over scientific names. Until that ruling is issued, the tooth retains its official status.

For paleontologists who study small theropod dinosaurs, the implications extend well beyond taxonomy. Growth series of this quality are vanishingly rare in the fossil record, particularly for animals as small and lightly built as troodontids, whose delicate bones are less likely to survive fossilization than those of their larger cousins. Having a well-sampled embryonic-to-adult sequence of pelvic and hindlimb anatomy provides a reference framework that other researchers can use to identify isolated bones, assess growth stage in new discoveries, and test hypotheses about growth rates, locomotion, and life history across the broader troodontid family.

Future work is already being contemplated. The authors note that further research could illuminate how Troodon walked and ran—questions that depend critically on precisely the kind of limb-proportion and joint-architecture data this study provides. Understanding the biomechanics of an animal with proportionally long lower legs and a bird-like skeletal plan could offer fresh insights into the evolutionary transition toward the avian body plan and the range of ecological roles small feathered predators occupied in late Cretaceous ecosystems.

The research also underscores the continuing scientific value of long-term, careful fieldwork in well-studied formations like the Two Medicine. The fossils analyzed in this study were not the product of a single spectacular discovery but of accumulated collecting across decades, in an area where the same troodontid population left behind eggs, embryos, juveniles, and adults in close geological proximity. It is precisely this kind of cumulative evidence that allows paleontologists to move beyond naming new species and toward reconstructing the lives of animals that have been extinct for 75 million years.

As the authors themselves put it, the analysis of these bones “provided an unprecedented opportunity for understanding the biology of a historically enigmatic group of animals and will serve as a valuable resource for paleontologists for years to come.” Whether or not the International Commission approves the proposed type-specimen change, the study stands as a landmark in the rehabilitation of a dinosaur that science has known longest by its teeth—and is only now beginning to know by its bones.

The research reconstructs environmental conditions across two periods of major instability separated by more than 100,000 years. The first is the last interglacial, known in the paleoclimate record as Marine Isotope Stage 5e, which lasted from roughly 129,000 to 116,000 years ago. The second is the early Holocene, between approximately 11,000 and 7,000 years ago, when agricultural communities were becoming firmly established across the region. To compare these intervals, Frumkin drew together an unusually broad range of evidence: fine charcoal particles preserved in ancient lake sediments that record peaks in regional burning, chemical signatures archived in cave formations that document substantial losses of vegetation and soil, and the distribution of archaeological sites that show where people settled as the landscape changed. Together, these records indicate that both periods featured episodes of extensive wildfire followed by severe erosion, with the most intense Holocene episode occurring between roughly 9,500 and 8,200 years ago, precisely the window in which the earliest pottery appears in the Levant.

The mechanism linking fire to landscape change is straightforward geomorphology, but its consequences were profound. When intense fires removed the vegetation cover that normally anchors hillslope soils, the exposed ground became vulnerable to rainfall and runoff. In the southern Levant, much of that soil was terra rossa, a reddish, clay-rich earth that develops over limestone bedrock. Once stripped of protective plants, hillsides surrendered this material to the rains, and it was swept into valleys and depressions, including the Jordan Valley and the Dead Sea basin, where it accumulated in deposits that can reach spectacular thicknesses. Far from merely destroying habitable land, this process created new environments where farming could flourish. Large Neolithic settlements increasingly clustered on these fertile redeposited soils, meaning that the very disaster that burned the hills also concentrated people and clay side by side in the valleys below.

It is that juxtaposition of people and freshly deposited clay that lies at the heart of the natural kiln hypothesis. Pottery emerged in the region during roughly the same broad period in which fires, erosion and clay deposition reached their peak. Frumkin proposes that people living in the valleys may have encountered patches of clay that intense wildfires had baked in place, and noticed that heat had transformed the soft, plastic material into something hard and durable. Wildfires burning through accumulations of dry brush, woody debris and clay-rich sediment can generate temperatures in the immediate ground layer that approach the threshold required for ceramic transformation. Observing such naturally fired clay could have provided the conceptual leap behind intentional pottery making: the realization that a process seen in the aftermath of a fire could be reproduced deliberately, under controlled conditions, at the hearth or in a pit.

The hypothesis gains support from earlier petrographic research on some of the region’s oldest ceramics. Thin-section analysis of Yarmukian pottery, one of the earliest ceramic traditions of the southern Levant, shows that these vessels were manufactured from local redeposited terra rossa clay, the very material that would have accumulated around settlements as the surrounding hills eroded. In other words, the earliest potters of the Jordan Valley were already working with the exact raw material that fire and flood had delivered to their doorstep. That correspondence between the geological record of fire-induced erosion and the mineralogical fingerprint of the earliest pottery is precisely the kind of pattern one would expect if environmental disruption and ceramic innovation were genuinely linked, rather than coincidentally adjacent in time.

Perhaps the most intriguing element of the study is a kind of natural experiment run more than a hundred millennia before pottery appeared. During the last interglacial, the southern Levant experienced an even more extreme combination of fires, vegetation loss, erosion and clay deposition than the early Holocene. Yet pottery did not emerge. Frumkin treats this earlier period as a control case: the environment produced many of the same ingredients, but the people of the time lacked the cognitive capacities, sedentary agricultural society, accumulated technological knowledge and social conditions that later Neolithic communities possessed. The comparison suggests that wildfire and clay deposition alone were not sufficient to produce ceramics. Innovation, in this view, required the meeting of environmental opportunity with a society capable of recognizing and exploiting it. Nature could stage the demonstration; only a prepared audience could learn from it.

The broader significance of pottery is difficult to overstate. Once fired vessels entered daily life, they transformed it. Ceramic containers made it far easier to cook, store food and liquids, and ferment products, supporting larger, more settled populations and reshaping economies built around surplus and storage. Decorated vessels and figurines quickly took on cultural and symbolic importance as well, indicating that ceramics acquired meaning far beyond their practical uses almost immediately after their invention. If Frumkin is right, the chain of causation running from climatically driven wildfires, through erosion and valley-floor clay deposition, to the rise of settled communities and finally to the invention of pottery, represents one of the most striking examples in human history of catastrophe opening a door to innovation.

Frumkin is careful to frame the natural kiln hypothesis as speculative and testable rather than proven. The available evidence establishes that major fires, clay deposition and the emergence of pottery occurred within the same broad period, but current dating resolution cannot yet demonstrate the exact sequence of events. He outlines concrete ways the idea could be examined. Future excavations beneath the earliest pottery-bearing layers could search for naturally fire-baked clay, which would constitute direct physical evidence that nature had demonstrated the process before people replicated it. Experimental research could test whether wildfires in the Jordan Valley could realistically heat clay to the roughly 500 to 700 degrees Celsius needed to begin permanently transforming it into ceramic material, by reconstructing fuel loads, burning conditions and the thermal properties of terra rossa deposits. And, as Frumkin notes, the hypothesis faces a clear potential refutation: the discovery of pottery that unambiguously predates the major fire episode would undermine the claim that the fires played a role in the invention.

The proposal does not seek to replace existing explanations for the origin of ceramics. Archaeologists have long argued that pottery emerged through gradual experimentation, driven by social needs such as cooking and storage, or spread through contact between communities. What the natural kiln hypothesis adds is another possible piece of the puzzle, one that places environmental agency back into a story usually told in purely cultural terms. It suggests that environmental catastrophe itself may sometimes create the conditions for technological breakthroughs, not by destroying societies, but by exposing them to new materials, new landscapes and new possibilities they would otherwise never have encountered.

In the end, the study paints a striking picture of the deep past: a Levant alternately scorched and flooded, its hillsides burning and its valleys filling with clay, while farming communities adapted, clustered and ultimately prospered on the wreckage. The fires that stripped those ancient hills may also have reshaped where people lived, how they farmed and, potentially, how they first learned to turn ordinary earth into one of civilization’s most enduring materials. If future fieldwork and experiments bear out Frumkin’s idea, the invention of pottery will join a short and remarkable list of technologies that humanity did not so much invent as learn to imitate from the world around it, after watching nature perform the trick first.

The fragments in question, often described in the literature as “debitage,” are the crumbs of sarsen — a hard, silicified sandstone — left behind or displaced during the monument’s construction and later disturbance. While the towering trilithons have attracted centuries of attention, these palm-sized chips have long been regarded as geoarchaeological afterthoughts. That changed when previous researchers applied geochemical provenancing to the fragments, measuring their elemental compositions and comparing them with known sarsen sources across southern Britain. That earlier work concluded that most of the stones, including the massive sarsen megaliths themselves, could be traced to the Marlborough Downs, roughly 25 kilometers north of Stonehenge, while some fragments pointed to more distant or exotic origins.

Michelaki and her colleagues — including David Barham, Michael P. Gorton, William C. Mahaney, Susanne Aufreiter, and Ronald G. V. Hancock — did not collect new samples. Instead, they took the previously published datasets and subjected them to a different kind of analysis. The crux of their critique concerns normalization, a widely used technique in geochemistry in which the concentration of each element is divided by the concentration of a reference element — in this case zirconium, or Zr. Normalization is intended to strip away noise caused by dilution effects, such as varying amounts of quartz in a rock, allowing researchers to compare the underlying chemistry of samples on a level playing field.

The technique has deep roots. Archaeometric studies of pottery, sediments, and metals have long relied on normalizing elements — aluminum for estuarine sediments, scandium for ceramic pastes, zirconium for silica-rich rocks — to counteract the enormous spreads in raw elemental concentrations that obscure meaningful comparisons. In the Stonehenge work, Zr normalization formed the backbone of the provenance assignments: when fragment ratios matched the ratios of a candidate source, the fragment was deemed to originate there. On the surface, the logic is sound and elegant.

But the new reanalysis shows that the elegance comes at a cost. When the same data are plotted without normalization, using raw elemental concentrations and a visual, geochemically informed graphical approach that explicitly acknowledges analytical variability, a different picture emerges. Similarities that appeared compelling under Zr-normalized ratios are not always supported across multiple elements. In some cases, normalization generated apparent agreements between fragments and sources that dissolve the moment other elements are considered. In other cases, genuine coarse geochemical groupings visible in the raw data simply vanish when everything is divided by zirconium.

The problem, the authors explain, is mathematical as much as geological. Dividing one element by another collapses two independent pieces of information into a single ratio, and if the normalizing element itself varies substantially across samples — as Zr can in silcretes, where zircon-bearing grains may be unevenly distributed — the ratio can swing wildly for reasons that have nothing to do with shared provenance. Two chemically unrelated rocks can end up with similar ratios, while two chemically related rocks can be driven apart. The result is a classification scheme that may obscure real compositional differences and invent apparent ones.

So what happens to the Stonehenge fragments when the crutches are removed? Using their non-normalized approach, Michelaki’s team found that only some of the fragments can be tentatively associated with previously characterized sarsen sources. Many others remain stubbornly unsourced, or appear to represent geochemistries that have not yet been documented anywhere in Britain’s known sarsen outcrops. The authors are careful with their language — they use “grouping” rather than “group” to stress that these chemical clusters are far more dispersed than the tight compositional families typical of a single geological source. But the implication is striking: the geochemical variability of the silcrete materials at Stonehenge may be considerably greater than the earlier analyses suggested.

Far from undermining the broader narrative of Stonehenge’s construction, this increased variability actually aligns with a growing body of evidence that the monument is a composite of materials drawn from across Britain. The most dramatic example is the Altar Stone, whose recent sourcing to the Orcadian Basin in northern Scotland — hundreds of kilometers from Wiltshire — stunned the archaeological community and demonstrated that long-distance stone transport, whether by human effort or glacial action, was part of the monument’s biography. The bluestones from the Preseli Hills of Wales tell a similar story. Against this backdrop, sarsen fragments with unassigned or unexpected chemistries are less an anomaly than a further hint that the builders drew on a wider and more varied portfolio of stone than once assumed.

The study is also notable for what it declines to do. Rather than offering new definitive provenance assignments, the authors frame their contribution as a methodological intervention: a reflexive evaluation of how analytical choices shape archaeological interpretation. The Stonehenge case becomes a cautionary tale for provenance studies in general. Archaeometry has a long history of these debates, from the Olmec controversy, where competing statistical treatments of neutron activation data produced radically different pictures of ceramic exchange in ancient Mesoamerica, to ongoing discussions about how to define chemical reference groups in pottery. The lesson, repeatedly learned and just as repeatedly forgotten, is that data do not interpret themselves — and that a technique convenient enough to become standard practice deserves periodic scrutiny.

The timing of the intervention is significant. The original provenancing of the sarsens, published in Science Advances in 2020, was widely celebrated as solving one of Stonehenge’s enduring mysteries, and subsequent work on the fragments extended that framework. A parallel debate has been playing out in the pages of the journal Archaeometry over the sourcing of sarsen stone 58, with the present team contributing geochemical analyses of that monolith and engaging in exchanges with other researchers over methodology. The new paper effectively widens the conversation: if normalization can obscure differences and manufacture agreements in the fragment data, then provenance claims built on those ratios — for the fragments and potentially for the megaliths themselves — warrant reexamination under multiple analytical lenses.

None of this means the Marlborough Downs connection is wrong. The 2020 study’s core insight, that the great sarsens share a distinctive chemistry matching a specific outcrop called West Woods, was based on a robust correspondence and remains the most parsimonious explanation for the megaliths. But the new work insists that confidence should be calibrated to the evidence. Some fragment-to-source matches may hold up under non-normalized scrutiny; others may not; and a substantial portion of the material may simply await the characterization of sarsen sources that have not yet been sampled. Britain’s sarsen distribution is patchy and imperfectly mapped, scattered in a belt from Dorset through Wiltshire to Kent and beyond, and glacial processes during the last British-Irish Ice Sheet’s advance and retreat have redistributed stone across the landscape in ways that complicate simple local-versus-exotic dichotomies.

For the wider field, the stakes extend well beyond one monument. Provenance studies underpin claims about trade networks, migration, territoriality, and social organization from the Neolithic to the historic period, and they increasingly feature in heritage policy debates about stone conservation and sourcing. When a single statistical transformation can tip the balance between “local stone” and “stone from nowhere we know,” the responsibility on practitioners is considerable. The authors’ recommendation is not to abandon normalization outright — it remains a legitimate and often useful tool — but to apply it reflexively, testing whether conclusions survive when the data are viewed in their raw form, element by element.

In the end, the study is a portrait of science working as it should. Two teams, using the same numbers, reach different conclusions, and the disagreement itself becomes a discovery: the stone fragments of Stonehenge are more chemically diverse, and their journeys more complicated, than anyone had asserted. For a monument that has resisted definitive explanation for a thousand years of inquiry, that may be fitting. The researchers, who note their work was enabled by “curiosity, stubbornness, and retirement,” have reminded the field that at Stonehenge, as in geochemistry, certainty should always be earned — and re-earned — rather than assumed.

The sample comprised 297 non-adults and 793 adults, including 258 males, 244 females and 291 individuals of undetermined sex. Sex was estimated from the morphology of the innominate bones, cranium and mandible using standard anthropological criteria, applied only to individuals older than 15 because secondary sexual characteristics develop after puberty. Age at death in adults was reconstructed through a battery of well-established skeletal indicators, including the pubic symphysis, auricular surface of the ilium, acetabulum, ectocranial suture closure, sternal rib ends, the sternal end of the clavicle and patterns of dental wear. For non-adults, the researchers relied on the formation and fusion of ossification centers, the diaphyseal lengths of long bones and tooth eruption sequences. Individuals with severe skeletal pathologies capable of distorting body proportions, such as rickets, were excluded from stature and body mass analyses, whereas those with milder conditions such as osteoarthritis were retained.

Life expectancy at birth was calculated using a stationary population life table based on Halley’s model, an approach the authors selected because the burial assemblages reflect attritional rather than catastrophic mortality and because the method permits direct comparison with a substantial body of published paleodemographic research from the Iberian and Italian peninsulas. Stature was estimated with Pearson’s regression equations, derived from the physiological lengths of the left femur and tibia, chosen because the French reference sample is biologically and geographically the closest match to Mediterranean populations and has been shown to perform well on Spanish and Italian skeletal series. Body mass was derived from the femoral head diameter using equations developed by Ruff and colleagues, a well-validated proxy because the femoral head remodels in response to weight-bearing loads. Body mass index was then computed as body mass divided by the square of stature, with individuals classified as underweight, normal weight, overweight or obese according to conventional BMI thresholds. Statistical comparisons between the sexes employed Student’s t-tests or Mann-Whitney U tests depending on the distribution of the data, while temporal trends were tested with one-way ANOVA followed by Tukey post-hoc comparisons, all conducted in SPSS at a significance threshold of p < 0.05.

The results reveal a striking asymmetry between the four biological indicators. Life expectancy at birth remained relatively stable across most periods, hovering between roughly 21 and 27 years, before showing a marked increase during the Modern period. The Prehistoric value of just over 12 years is exceptionally low, but the authors caution that the small Prehistoric sample of 45 individuals, shaped by funerary practices such as adult cremation in the Late Bronze and Iron Ages, likely depresses this figure far below any biological reality. The Roman period yielded a life expectancy of about 21.6 years, squarely within the range documented for ancient Mediterranean populations, where nearly half of all children may have died before age five. Life expectancy peaked at approximately 27.2 years in Late Antiquity, dipped to about 25 years during the turbulent Medieval centuries, and climbed again to 27.15 years in the Modern period.

Stature, by contrast, displayed no statistically significant variation across the seven millennia under study. Male averages ranged narrowly between 164 and 166 centimeters, and female averages between 151 and 154 centimeters, in every period analyzed. Males were significantly taller than females throughout, as expected from sexual dimorphism, but the constancy of absolute height across such dramatic cultural transformations, from Neolithic farming villages to the proto-industrial city, surprised the researchers. The finding parallels a comparable diachronic study of Milan, and the authors suggest that restricting the analysis to a single urban area minimizes the confounding effects of geography, microclimate, agricultural systems and economic activity that have produced the U-shaped stature trends reported in earlier studies that mixed rural and urban, and multi-regional, samples.

Body mass and body mass index told a very different story, showing statistically significant changes over time in both sexes. In the Prehistoric assemblage, virtually all individuals fell within the normal weight range, with only about 14 percent of males classified as overweight and no underweight or obese individuals at all. Overweight cases rose during the Roman period to roughly 30 percent of males and 29 percent of females. Obesity made its first appearance in the archaeological record of Barcelona during Late Antiquity, detected in about 1 percent of females, alongside the earliest underweight individuals. During the Medieval period, male body mass dropped to its lowest recorded average of 59 kilograms, and underweight classifications climbed to over 19 percent of males, a pattern the authors link to famines, warfare, sieges and epidemic disease that plagued the city between the ninth and fourteenth centuries.

The Modern period brought a decisive reversal. Body mass peaked in both sexes, and overweight prevalence surged to 47 percent of males and 40 percent of females, the highest levels in the entire sequence. The authors attribute this shift to profound dietary and lifestyle transformations accompanying the transition from the pre-industrial to the industrial era: the Columbian Exchange introduced American crops such as maize and potatoes into the diets of the poor, proto-industrialization and textile manufacturing spurred economic recovery by the eighteenth century, and Spain experienced declining mortality thanks to improved nutrition, hygiene and disease treatment. Notably, the estimated body masses of Modern-period Barcelonans, averaging 67.6 kilograms for males and 59.5 kilograms for females, closely match those of post-medieval Dutch urban populations from Arnhem and Zwolle, underscoring a shared European trajectory of increasing body size in urban settings.

The study’s demographic data also captured the scars of Barcelona’s most catastrophic episodes. The Medieval decline in life expectancy, body mass and BMI coincides with centuries of conflict along the Spanish March between Islamic and Christian forces, recurrent famines, the devastating sack of the city by Almanzor’s army in 985, in which much of the population was killed or enslaved, and repeated outbreaks of the Black Death in the fourteenth century that decimated the population by as much as 40 percent. The city required roughly two centuries to recover its pre-1348 population levels. Violence against marginalized communities, including the destruction of the Jewish neighborhood of El Call, further marked this grim chapter. These historical events, the authors argue, left measurable imprints on the biological record, demonstrating that social inequality, conflict and epidemics translate directly into skeletal evidence of stress and mortality.

A central conclusion of the work is that life expectancy, body mass and body mass index proved far more sensitive to sociocultural and environmental change than stature did. While the skeleton preserved height with remarkable stability, the other three parameters registered three major shifts: the transition from Prehistory to the Roman period, a pronounced downturn during the Medieval era, and a strong recovery between the Medieval and Modern periods. Because the necropolises sampled predominantly contain individuals of lower socioeconomic status, buried in simple earth-cut graves rather than costly chapel tombs, the dataset offers a consistent window into the living conditions of ordinary working Barcelonans, from skilled tradesmen such as carpenters and weavers to unskilled laborers, across nearly seven millennia.

The authors acknowledge the inherent limitations of paleodemographic reconstruction, including the “osteological paradox,” the chronological gap between the Neolithic and Bronze Age assemblages, and the scarcity of comparable body mass data from other historical Spanish populations. Nevertheless, the study establishes a foundational reference framework for future diachronic bioarchaeological research in the Iberian Peninsula and contributes to a broader scientific conversation about how human biology, environment and history intertwine. It also carries a contemporary resonance: obesity, once entirely absent from the city’s dead, now ranks among its most pressing public health challenges, and the skeletal record makes clear that this epidemic is a product of the modern era’s abundance and sedentary living, not a feature of the human condition.

The study, led by Sowmeya Sathiyamani and Claudia Colini of the Centre for the Study of Manuscript Cultures at Universität Hamburg, together with colleagues from Radboud University and several European institutions, examined fifty-six papyrus fragments bearing sixty-five independent texts, dated between the fifth and tenth centuries CE. The manuscripts came from five collections across Europe: the Austrian National Library in Vienna, the Institute of Papyrology of Sorbonne University, the Michaelides collection at Cambridge University Library, Leiden University Library, and the State and University Library in Hamburg. Crucially, the team employed an entirely non-invasive methodology, meaning none of these fragile artifacts was damaged or sampled during analysis—a critical consideration when dealing with materials that have survived more than a thousand years.

The technical backbone of the investigation rested on three complementary techniques. First, two-colour digital microscopy using a Dino-Lite USB microscope operating at magnifications of forty to fifty times allowed researchers to image ink strokes under both visible light and short-wave near-infrared illumination at 940 nanometers. This provided an initial diagnostic signature: carbon inks retain their absorptivity and remain opaque under near-infrared light, while iron-gall inks lose opacity, and pure plant inks become completely transparent. Second, infrared reflectography was performed with an OPUS Apollo imaging system equipped with a cooled indium gallium arsenide sensor sensitive in the 900 to 1700 nanometer range. Using a long wave-pass filter restricting detection to 1510–1700 nanometers, the researchers could definitively distinguish ink types: pure iron-gall ink becomes fully transparent at around 1400 nanometers, whereas any ink containing carbon—including mixed formulations—continues to absorb radiation and remains legible. Images were processed with a custom MATLAB script and normalized against a white Labsphere Spectralon diffuse reflectance target reflecting ninety-nine percent of light across the sensor’s range. Third, X-ray fluorescence spectroscopy provided elemental fingerprints of the inks. The team used a Bruker CRONO spectrometer with a 10-watt rhodium X-ray tube and a 50-square-millimeter silicon drift detector, as well as a Bruker M6 Jetstream micro-XRF instrument with polycapillary optics and spot sizes as small as 30 micrometers, operating at voltages between 35 and 50 kilovolts. Rather than relying on single-point measurements, the researchers favored mapping acquisitions, which account for the notoriously heterogeneous nature of papyrus as a writing support. The resulting elemental distribution maps were rendered as heat maps in which red indicated the highest signal intensity and dark blue the lowest.

The results for the papyrus protocols were astonishingly uniform. All twenty-four Arabic and Arabic-Greek bilingual protocols, spanning the seventh through tenth centuries CE, were written with iron-gall ink containing iron, accompanied in most cases by potassium—attributable to the tannins and gum arabic used in ink manufacture—and in roughly half the samples by manganese, an impurity associated with iron. Notably, copper and zinc sulphates, which typically appear as satellite impurities when commercial iron sulphate salts are used, were absent from every Arabic protocol analyzed. Sulphur was detected in only eleven of the twenty-four fragments. This chemical profile implies either the deliberate use of an unusually pure iron sulphate source—perhaps the green or yellow vitriol referenced in Arabic recipes—or the preparation of ink from metallic iron, such as nails or filings, reacting with a vitriolic solution. The researchers connect this observation to the Arabic ink-making tradition: the Persian physician and alchemist Muḥammad ibn Zakariyyāʾ al-Rāzī, writing his Kitāb al-asrār (Book of Secrets) around 900 CE, identified six vitriolic substances and described purification methods, and later Arabic treatises transmit recipes combining iron nails with vitriol—chemistry that would naturally yield an iron-gall ink containing only iron, sulphur, and traces of manganese.

To place the Arabic protocols in historical context, the team also analyzed seven Byzantine protocols dating to the fifth and sixth centuries. Here the results revealed a temporal split with profound implications. The early Byzantine protocols, dated between the early fifth and mid-sixth centuries, were written with carbon-based ink—a conclusion confirmed for three fragments by supplementary Raman spectroscopy using a 532-nanometer laser when the infrared reflectography camera was unavailable. The later Byzantine protocols, from the mid-sixth to late seventh centuries, however, employed either a plant ink or an iron-poor iron-gall ink, with XRF mapping detecting iron, potassium, and traces of copper. This means the Arabic protocols, though their textual formulas derive from the earlier Byzantine tradition, followed the ink practice of the later Byzantine era. When the Arab conquest transferred ownership of Egypt’s papyrus manufacturing facilities to the new rulers, the incoming administration adopted not merely the bureaucratic system of certifying rolls with protocols but the very ink chemistry associated with that function.

Papyrus protocols occupy a unique position among documentary texts, and understanding why illuminates the significance of the finding. A protocol is the text written on the prōtokollon, the outermost sheet of a papyrus roll, certifying its production in a state-controlled facility. Introduced in the Byzantine period around the fifth century CE and mandated as proof of authenticity for notarial documents by Justinian’s decree of 537 CE, the protocol evolved from a few lines in the center of the first sheet to text covering nearly the whole surface, including nearly illegible perpendicular writing known as Schraffenschrift. The bilingual protocols transitioned to purely Arabic text around 720 CE, likely as part of an empire-wide adoption of Arabic as the official administrative language, and Qur’anic verses became common elements in the formula by the second half of the eighth century. Because protocols were necessarily tied to a single writing support and produced in a localized industry—predominantly in the Nile Delta—they form a coherent corpus united by function and context of production, making them ideal subjects for correlating material choice with documentary purpose.

The contrast with other document categories was equally telling. Nine official letters from the dossier of Qurra ibn Sharīk, governor of Egypt from 709 to 714 CE, written in both Greek and Arabic and dispatched from the chancellery at al-Fusṭāṭ to Basilios, pagarch of Aphrodito, were all written with carbon-based inks. Interestingly, when metallic impurities appeared, they correlated with language: copper was detected mostly on the Arabic papyri, while iron appeared mostly with the Greek texts. Combined with the known palaeographic evidence that the Arabic letters were written with a broad-cut reed pen and the Greek with a narrow-cut pen, this suggests that distinct inks, inkwells, and likely scribes were used for each language within the same chancellery—a hypothesis consistent with the staged production of these bilingual documents. Thirteen further administrative and legal documents, including contracts of sale, deeds of donation, and receipts, along with twelve private and religious texts, were also written with carbon-based inks, irrespective of their degree of formality. The single exception was a private letter, P.Hamb. Arab. 6, dated 927–928 CE, written with a carbon–iron-gall mixed ink, detectable because the ink lost opacity under infrared reflectography while XRF revealed calcium, titanium, manganese, and iron.

The researchers propose several overlapping explanations for this functional division. One factor may be accessibility and cost: producing high-quality soot for carbon ink is labor-intensive and requires specialized facilities, yet lower grades could be mass-produced and distributed, whereas iron-gall inks are easy to make in small batches once raw materials—iron sources and tannins, often imported oak gall nuts—are obtained. A second, arguably more compelling explanation lies in the physical chemistry of the inks themselves. Iron-gall ink penetrates the writing support through the formation of an iron(III)-polyphenol complex, whereas carbon ink largely sits on the surface, making iron-gall text considerably harder to erase. For protocols serving as certificates of authenticity under a state monopoly—where Justinian’s decree required the protocol-bearing sheet to be attached to legal documents—permanence against tampering was paramount. Arabic erasure recipes compiled centuries later confirm that removing iron-gall writing is feasible only under mildly acidic conditions. In contrast, the documentary letters of a governor, though official, had limited temporal validity and did not demand the same permanence, making convenient, erasable carbon ink the practical choice.

Perhaps the most striking demonstration of the principle came from a direct comparison: two bilingual protocols, P.Vindob. G39766 and P.Vindob. G39767, were produced during the very same governorship of Qurra ibn Sharīk as the carbon-ink chancellery letters, yet the protocols were written in iron-gall ink while the letters used carbon ink. The choice of ink, the authors conclude, was not determined by the person in power but by documentary traditions that predated the Islamic conquest of Egypt. The study thereby provides material evidence of how pre-Islamic scribal conventions shaped the development of Arabic administrative culture, showing that the chemical signature of an ink can serve as a witness to institutional continuity across the Byzantine-Islamic transition. The authors caution that their corpus, while broad, remains small relative to the thousands of extant protocols—the analyzed Arabic protocols represent roughly five percent of catalogued examples from the seventh to tenth centuries—and that qualitative XRF data carry inherent limitations, including the difficulty of detecting low-atomic-weight elements like sulphur. Even so, the work demonstrates how non-invasive archaeometry can decode the silent, deliberate choices of scribes whose recipes were never written down, transforming faint brown strokes on yellowed papyrus into a readable record of technological tradition, bureaucratic logic, and cultural inheritance.

Ethnoarchaeology, the branch of the discipline that studies living communities to generate analogies for interpreting the material remains of past societies, has long occupied an uneasy position on the continent. Researchers arrive with notebooks and research permits, document pottery production, house construction, or trade practices, and then leave—often without returning results, sharing credit, or addressing community priorities. The authors describe this extractive model as increasingly untenable, particularly in a moment when the African Union has declared 2025 a year focused on reparations and when demands for the restitution of African cultural heritage have accelerated worldwide. Against this backdrop, the Amedeka Archaeological Project represents something more ambitious than a fieldwork strategy: it is an attempt to build durable, reciprocal relationships in which community members are collaborators rather than subjects.

Amedeka is a Krobo community in the Lower Manya Krobo area of Ghana’s southeastern region, situated near the Volta River and within the broader landscape transformed by the Akosombo Dam, completed in the 1960s as the centerpiece of the Volta River Project. The dam’s construction flooded vast territories and resettled thousands of people, and its downstream effects—altered fisheries, changing floodplain agriculture, and, most recently, the catastrophic spillage of 2023 that displaced communities downstream—have made the region a living laboratory of environmental and social change. The authors situate Amedeka’s past within this layered history, tracing how 19th-century trade networks, the colonial commercial transition, and 20th-century state development schemes shaped what people bought, sold, consumed, and discarded. Kuma’s earlier research in the same community, published separately, reconstructed trade and consumer practices from the late 19th to mid-20th centuries, including the local negotiation of imported goods known as “Nkudzedze,” while companion work documented the spatio-temporal distribution of locally produced earthenwares.

The methodological core of the project is what the authors call community-based collaboration, and it operates on several levels simultaneously. Elders and knowledge holders in Amedeka participate in the interpretation of archaeological and oral historical data, not merely as informants but as analysts whose frameworks—Krobo concepts of personhood, community, and historical memory—are treated as epistemologically valid. Fredrick Tekpor, a resident of Amedeka and co-author of the paper, exemplifies this structure: his intellectual contribution is credited not as assistance but as authorship. Data governance follows the same logic. The study’s data availability statement notes that materials sensitive to the community are subject to access restrictions established through agreements with the Amedeka community, a mechanism that operationalizes the principle that communities can refuse, restrict, or condition research outputs concerning their own heritage.

The authors frame the future of equitable research around three interconnected dimensions. The first is positionality: an honest accounting of who the researchers are, where they stand relative to the community, and how their identities—Ghanaian, diasporic, institutional—shape what gets asked and what gets heard. Kuma and Gavua, both Ghanaian scholars working across North American, Ghanaian, and European institutions, reflect on the way their own embeddedness in academic systems of evaluation, funding, and publication creates pressures that can conflict with community timelines. The second dimension is the evolution of research questions themselves. Rather than fixing questions in a proposal written in Pittsburgh or Accra, the project allows questions to shift in response to community concerns—including, the authors suggest, issues of land access, heritage loss, and the community’s desire to benefit from tourism development along the Volta. The third dimension is partnership: the slow, sometimes fragile work of building institutions of accountability that survive beyond a single field season or grant cycle.

That slowness is deliberate, and the authors lean on a growing body of scholarship that treats it as a methodological virtue rather than a inefficiency. Cunningham and MacEachern’s influential argument that ethnoarchaeology should be understood as “slow science” is central here, as is Stahl’s call for assembling “effective archaeologies” oriented toward equitable futures. The Amedeka project also draws on Indigenous archaeology movements in North America and Oceania, on Black feminist and antiracist archaeologies, and on African philosophical frameworks—most notably Ubuntu, the relational ethic that locates personhood and knowledge in community. By citing these traditions together, the authors make a technical point about method: collaborative research is not simply conventional archaeology plus politeness. It requires different temporal rhythms, different consent structures, different instruments of data sovereignty, and different metrics of success than those embedded in the grant-and-publication machinery of the neoliberal university.

The practical stakes are considerable. Ghana’s legal framework, under the Environmental Protection Act and its predecessors, mandates environmental impact assessments for projects likely to have adverse environmental effects, yet archaeological and heritage considerations routinely enter such processes late, shallowly, or not at all. Previous large infrastructure projects in Ghana, including the Bui Dam, generated documented conflicts in which communities declared they would not relocate until their ancestors and shrines came with them—a signal that heritage is not an ornament of development but a condition of its legitimacy. The Amedeka paper argues that community-engaged archaeology, done early and done well, can function as an alternative archive: a way of surfacing local histories, sacred landscapes, and material claims before planning decisions harden into displacement. This echoes work by Logan, who showed how archaeology at Banda, Ghana, can serve as an archive of food security, and by Boateng and colleagues, who used archaeological knowledge of Africa’s past to inform safer food-packaging alternatives in Accra.

The authors are candid about the difficulties. Community archaeology in Africa has a history of what Chirikure, Pwiti, and others called unfulfilled promises: participatory rhetoric that dissolves into one-off consultations, museum-driven agendas, and tourism revenue that never reaches residents. Genuine co-production, the paper argues, requires that communities be able to shape research design from the beginning, that findings be returned in accessible and locally useful forms, and that funding structures—including the National Geographic Society grant and Social Science Research Council support that funded this work—be flexible enough to accommodate community-paced timelines. It also requires what the authors, following scholars such as Borrows, Tuck and Yang, and Santos, describe as humility and even refusal: the willingness of researchers to accept that some knowledge is not theirs to collect, that some questions will be redirected, and that the epistemologies of the South are not raw material for Northern theory but complete systems in their own right.

What emerges from the paper is less a template than a demonstration that the model can work. The Amedeka Archaeological Project has generated rigorous archaeological data—on earthenware production and consumption, on trade networks, on household material culture—while simultaneously producing a community of co-researchers, a set of enforceable data agreements, and a publication whose author list includes the community itself. Earlier versions of the work were presented at the Society of Africanist Archaeologists’ biennial meeting, and the project’s public-facing engagement has included recorded community events shared online, signaling a commitment to communicating results back to the people who made them possible.

For the wider discipline, the study arrives at a charged moment. Calls for an antiracist archaeology, for restitution of looted and removed African heritage, and for decolonized research methodologies have moved from conference margins to institutional agendas. The Ghanaian government’s Ministry of Tourism, Culture and Creative Arts has publicly reinvigorated restitution efforts, and the African Union’s reparations framing has given continental political weight to what were once framed as academic concerns. Within that landscape, the Amedeka paper’s central claim is quietly radical: that the future of African archaeology will be judged not only by what it discovers, but by whether the communities whose pasts are at stake gather around that past as authors of it—and whether the discipline is prepared to gather with them, on their terms, for the long term.

The study emerges from data generated by the Feeding Anglo-Saxon England project, an ambitious research initiative that catalogued thousands of animal bone assemblages from excavated settlements, farms, and towns. By pooling this information, Holmes and Orton were able to move beyond the fragmented site-by-site narratives that have long dominated medieval zooarchaeology. Instead of relying on the conventional periodization that divides the past into eras such as “Saxon” or “high medieval,” the team applied an aoristic dating method, a statistical technique that distributes the probability of an observation across the possible date ranges of its context. This allowed them to trace changes in livestock management along a continuous timeline, smoothing out the artificial steps imposed by archaeological convention and exposing trends that would otherwise be blurred or misplaced.

The first major shift the researchers identified concerns the rise of sheep. In the earliest centuries of the study window, roughly the fifth and sixth centuries, cattle dominated the animal economy of England. Bones of cows far outnumbered those of sheep at sites across the country, reflecting an agricultural system rooted in dairying, plowing, and beef. But beginning in the seventh century, sheep numbers began to climb, and the geographical pattern of that increase pointed to an unexpected instigator: the newly founded monastic communities of the late Anglo-Saxon period. Monastic farms, the authors argue, placed a deliberate emphasis on wool production, a fiber that could be exported, traded, and accumulated as wealth in ways that perishable meat and milk could not. What began as a strategy of religious estates rippled outward, and by the end of the medieval period wool had become the backbone of the English economy, a position it would hold well into the sixteenth century.

The second turning point involved cattle themselves, and it was one the researchers did not anticipate. Between the tenth and twelfth centuries, the age profile of male cattle in the archaeological record shifted markedly toward older animals. In zooarchaeology, the age at which livestock are slaughtered is a critical diagnostic: young culling points toward meat production, while the survival of animals well into adulthood suggests they were kept for other purposes, such as milk, breeding, or labor. The medieval bones told an unambiguous story of increasing numbers of elderly male cattle, and when the team combined this with the growing incidence of deformations to the lower limb bones of these animals, changes characteristic of the sustained strain of pulling heavy loads, the interpretation became clear. Farmers of the tenth to twelfth centuries were working their oxen harder and longer than ever before, deploying them as draught animals to power an expanding arable economy.

This peak in oxen use, the authors suggest, represents a genuine transformation in working practices rather than a mere technical adjustment. Keeping draught cattle alive and productive over many years would have demanded closer, more sustained relationships between farmers and their animals, with oxen functioning as valued members of the agricultural workforce rather than simply as commodities awaiting slaughter. The finding underscores how much of medieval economic history is written not in documents alone but in the wear patterns and healed injuries of animal skeletons, traces of labor that no charter or account book records.

Yet the age of the ox was not destined to last. After the twelfth century, the pattern reversed. Cattle were culled at younger ages, a signal that the market was demanding more meat and milk, and the role of draught cattle waned accordingly. Two forces appear to have converged to bring this about. The first was demographic: England’s population grew dramatically through the thirteenth and early fourteenth centuries, multiplying demand for food and reshaping the economics of livestock keeping toward products that could feed more people. The second was technological and practical: draught horses increasingly took over the labor previously performed by oxen, freeing cattle to be managed primarily as sources of beef and dairy. The combined effect redirected the animal economy from one organized around the energy output of cattle to one organized around their bodies as food.

The study’s continuous chronology clarifies the timing of these transitions in ways that earlier, period-bound analyses could not. The rise of sheep farming, for instance, can now be anchored firmly to the seventh century and linked specifically to monastic enterprise, pushing back the origins of England’s wool economy relative to some previous estimates. The fluctuating fortunes of draught cattle, rising across the tenth to twelfth centuries and declining thereafter, emerge as a discrete episode rather than part of a slow, linear progression. By seeing the animal economy as a continuous curve rather than a sequence of static snapshots, the researchers could detect both the acceleration and the reversal of trends, capturing the dynamism of a rural economy that was constantly renegotiating the balance between labor, food, and fiber.

The methodological innovation at the heart of the study, the aoristic approach to chronological uncertainty, may prove as influential as its historical findings. Archaeological contexts rarely yield precise dates; a bone from a pit may belong anywhere within a span of a century or more. Traditional analyses force such evidence into broad periods, sacrificing resolution and sometimes creating spurious transitions at period boundaries. By contrast, aoristic methods spread each observation probabilistically across its possible range, allowing the aggregate dataset to express genuine trends along a continuous axis. Holmes notes that there is no reason the approach cannot be extended to other classes of archaeological data, from ceramic assemblages to environmental indicators, potentially transforming how archaeologists interrogate large, imperfectly dated datasets.

The research also confirms and refines trends that had been reported in earlier scholarship, lending new confidence to a picture of medieval England as an economy in repeated flux. At the same time, the authors are candid about the limits of what their data can currently resolve. Significant variation exists between individual archaeological sites, reflecting local ecologies, estate structures, and market access, and much of this heterogeneity remains to be fully explored. Standards for determining the sex of ancient domesticated animals also vary across studies, an issue with real consequences for interpreting culling patterns and herd management, since the economic roles of male and female livestock differed profoundly. Both problems, the authors suggest, define an agenda for future work: finer spatial resolution and methodological standardization promise to sharpen the picture considerably.

The broader significance of the study lies in its demonstration that the great economic currents of medieval history, the growth of the wool trade, the intensification of agriculture, the rise of urban markets, left legible traces in the mundane debris of butchery and burial. The dataset confirms that animal husbandry in England was not a passive backdrop to human history but an active driver of it. Monastic decisions about flock management in the seventh century underpinned a national economy for nine hundred years. Farmers’ choices about how long to work their oxen in the tenth to twelfth centuries reshaped the arable landscape. And the pressure of a growing population in the thirteenth and fourteenth centuries rewrote the terms of trade between people and their livestock, prioritizing meat and milk over muscle.

For Holmes, who undertook the detailed analysis and interpretation after the conclusion of the European Research Council-funded FeedSax project that generated the underlying dataset, with support from her fellowship status at the University of Leicester, the work represents both an end and a beginning: a definitive statement about a millennium of English animal economy, and a template for analyzing change through time with methods matched to the realities of archaeological evidence. As excavations continue to add bones to the record and methods continue to mature, the continuous chronology pioneered in this study offers a way to hear, with increasing clarity, what the medieval countryside was telling us all along, one slaughtered ox and one shorn sheep at a time.

The authors have declared no competing interests. The underlying data collection was funded through the FeedSax project (ERC Horizon 2020), though the ERC played no role in the subsequent analysis, publication decisions, or preparation of the manuscript. The full article is freely available in PLOS One.

The research, led by Weihan Zou of Northwestern Polytechnical University and Sok Yee Yeo of Tongji University, together with colleagues at Xi’an Jiaotong University and Shanxi Ancient Architecture Engineering Supervision Co., was carried out ahead of conservation work on the temple. Because scientific studies of Song–Jin architectural paintings remain strikingly scarce, the team treated the surviving painted surfaces as a rare opportunity to document how early Chinese craftsmen decorated monumental timber buildings, and how later generations altered that decoration.

Architectural paintings occupy a distinctive place in Chinese building tradition. Unlike murals applied to walls, these are decorative schemes painted directly onto the structural members themselves — architraves, beams and brackets — following conventions recorded in imperial building manuals. Their polychrome patterns served both aesthetic and symbolic purposes, and their materials can reveal the technical knowledge, economic circumstances and workshop traditions of the period in which they were applied. But centuries of exposure, repair and renewal mean that what survives on a temple’s timbers today is rarely what was painted first.

To disentangle that history, the researchers collected twelve micro-samples from the architrave and beam paintings of the temple. From fragments often smaller than a fingernail, they extracted an extraordinary amount of information using a multi-analytical strategy in which each technique interrogates a different aspect of the material. Optical microscopy and cross-sectional analysis revealed the stratigraphy — the sequence of paint and ground layers stacked on top of one another. Scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM–EDS) mapped the elemental composition of each layer, identifying characteristic elements such as iron, copper, calcium and silicon. Raman spectroscopy and Fourier-transform infrared spectroscopy (FTIR) provided molecular fingerprints of pigments and organic binders, while X-ray diffraction confirmed crystalline phases. Finally, pyrolysis gas chromatography–mass spectrometry (Py–GC/MS) identified the organic binding media, the most elusive component of ancient paints.

The cross-sections told the first and most consequential story. The architrave paintings displayed a clear preparatory ground layer beneath the pigment layers, composed of gypsum (calcium sulfate), quartz and plant fibers — a deliberate recipe designed to create a smooth, pale surface on rough timber. Above this ground, multiple distinct pigment layers were visible, confirming that the architraves had been repainted on several occasions across different historical periods. The beam paintings, by contrast, lacked any preparatory ground at all; the pigments were applied directly to the wood. This fundamental difference in preparation suggests the two sets of paintings were not produced in the same campaign, by the same workshop, or even within the same century.

The pigment palette identified across the samples included iron(III) oxide — the familiar earthy red of hematite-based pigments — hydrated iron oxide, corresponding to yellow ochre-like materials, gypsum, amorphous carbon black, and malachite, a basic copper carbonate that supplied vivid green. SEM–EDS and Raman spectroscopy converged on these identifications: iron and oxygen signals paired with the characteristic Raman bands of hematite for the reds, copper signatures and carbonate bands for the malachite, and the broad, featureless Raman response of amorphous carbon for the blacks. Such a palette is consistent with mineral pigments available to painters in medieval China, drawn from locally obtainable ores and earths.

The binding media proved equally revealing. Py–GC/MS, which gently heats a sample and identifies the molecular fragments released, detected amino N-ethylpropanamide, palmitic acid and related pyrolysis products. Together, these markers point to a proteinaceous glue used in combination with walnut oil — a mixed binder system in which animal glue would have provided adhesion and working properties while the drying oil contributed cohesion and water resistance. The detection of specific fatty acids and protein pyrolysates in the same samples is the kind of molecular evidence that can distinguish, for example, a pure glue tempera from an oil-enriched formulation, and it places the Yuanrong materials firmly within documented Chinese painting practice.

When the team compared their results with published analyses of architectural paintings from other periods, a chronological pattern emerged. The beam paintings, with their ground-less preparation, appear to preserve an earlier technical tradition rooted in Tang-period practices, and the researchers suggest they may plausibly date to the early Song dynasty. The architrave paintings, with their gypsum-quartz-fiber ground and evidence of repeated repainting, show closer affinities with techniques documented from the Ming period, and may be linked to a later phase of repair or renewal. In other words, the temple’s decoration is not a snapshot but a palimpsest — earlier paint schemes surviving alongside, and beneath, later ones.

This finding carries implications well beyond a single building. Architectural paintings on surviving timber structures are often assumed, implicitly or explicitly, to date to the construction date of the building itself. The Yuanrong evidence demonstrates that the surviving paint can reflect different painting phases and successive renewals rather than a single original campaign. For architectural historians, this means that technical features such as the presence or absence of a ground layer, the choice of pigments, and the composition of binding media can serve as relative dating tools — complementing dendrochronology of the timbers and documentary records. For conservators, it means that any intervention must reckon with multiple, historically distinct paint campaigns coexisting on the same structural member, each with its own material vulnerabilities.

The study also fills a geographic and chronological gap. Most previous analytical work on Chinese architectural polychromy has concentrated on later imperial structures — Ming and Qing dynasty buildings in Beijing’s Forbidden City, imperial altars and Taoist temples — where abundant surviving decoration and institutional interest have driven research. Song–Jin period timber structures, of which Shanxi Province holds the greatest concentration, have received far less laboratory attention. By documenting the materials of a genuine Song–Jin building, the study provides a baseline against which other early wooden structures can be compared, helping to map the evolution of Chinese architectural painting technology from the Tang through the Song, Jin, Yuan and Ming periods.

The analytical workflow itself exemplifies modern heritage science practice. Micro-sampling — taking only tiny, strategically chosen fragments — minimizes harm to the historic fabric while enabling the full battery of instrumental techniques that would be impossible to apply non-invasively at the required sensitivity. Cross-sectional stratigraphy anchors the chemical data in physical space, so that each identified compound can be assigned to a specific layer and therefore a specific painting event. The combination of inorganic characterization (SEM–EDS, Raman, XRD) with organic analysis (FTIR, Py–GC/MS) is essential because architectural paintings are composite systems: their appearance depends on pigments, but their survival and behavior depend on grounds and binders. The discovery that walnut oil was used in a medieval Chinese architectural context is particularly noteworthy, as drying oils occupy a contested place in the history of East Asian painting, and every securely identified instance refines that picture.

For Yuanrong Temple itself, the results arrived at a critical moment. Conservation of ancient timber architecture typically involves cleaning, consolidation and, in some cases, repainting or protective coatings — decisions that can irreversibly alter or destroy evidence of earlier campaigns if made without analytical knowledge. By characterizing the stratigraphy and materials before treatment, the researchers have given conservators a map of what is preserved, where the boundaries between painting phases lie, and which layers carry the oldest surviving evidence. The beam paintings, potentially the earliest technical testimony in the building, can now be treated with the caution their age demands.

More broadly, the study is a reminder that buildings are documents. Every structure that has stood for a millennium has been touched, patched, and renewed by generations of caretakers, and each of those interventions left a material record — sometimes visible, often buried beneath later paint. Decoding that record requires the tools of analytical chemistry as much as the eyes of an art historian. At Yuanrong Temple, those tools have now shown that the reds of the architraves and the unground pigments of the beams belong to different conversations across time: one rooted in Tang tradition and the early Song, the other echoing Ming workshops centuries later. Together, they transform the temple from a single datable monument into a continuous, physical archive of Chinese painting practice, one microscopic layer at a time.

Feasting, defined by the researchers as ritualized collective consumption—the sharing of food among two or more people in an event marked as special—was far more than culinary indulgence in Swahili society. During the medieval Swahili period, roughly 800 to 1500 CE, stone towns emerged as political, economic, and ritual centers controlled by elites known as waungwana, who grew wealthy from regional and international trade and whose conversion to Islam connected them to transcontinental networks of believers. Food was central to their power. Historical accounts, including those of the fourteenth-century traveler Ibn Battuta, describe elaborate feasts used to entertain visitors and build influence. Nineteenth-century records describe families bankrupting themselves to supply enough rice, cattle, and goats for funerals, weddings, and circumcisions. Islam itself provided both new motives for feasting—weddings, funerals, and other life events framed within religious ideals—and a powerful interpretive framework, since feasts could be construed as acts of almsgiving fulfilling the obligation of zakat.

Mtwapa itself was an ideal setting for such displays. The walled town covered more than eight hectares and contained sixty-four stone houses, two cemeteries, and five mosques still standing when investigations began in 1986. Without a central administrative court like those at Gedi or Kilwa, the Friday Mosque Complex likely served religious, economic, and political functions simultaneously. Its layout reflected the hierarchical character of Swahili urbanism: an Eastern Gate led directly to the mosque and elite tombs, while a Western Gate gave access to the commoners’ cemetery and residential quarters. The complex featured extensive water infrastructure, including a large well connected by conduit to a cistern, or birka, supporting ritual purification for daily prayers and major communal events alike. Oral histories collected in 1991 from Mama Biimu, then the oldest resident and part-owner of Mtwapa, confirmed that the cemetery was reserved exclusively for the town’s elite—the Waungwana Wa Mtwapa, or Watwafi—while commoners were buried outside the walls.

The faunal remains analyzed in the study came from four excavation seasons spanning 1996 to 2011. The most important deposit was Locality 4, an eight-by-five-meter midden lying just beyond the mosque’s eastern gate. This midden accumulated inside a small stone guardhouse that had fallen into disuse and became a convenient dump site. Three radiocarbon dates place the deposits squarely in the fifteenth to seventeenth centuries. Several lines of evidence convinced the researchers that the material represented repeated feasting rather than ordinary refuse. Ethnohistorical accounts document ritual meal preparation and communal consumption to honor departed kin among Kenyan Swahili and Mijikenda communities. The sheer volume of faunal and ceramic remains was striking, and crucially, the ceramics consisted of local single-use coarse ware pottery lacking the smoke-stain patina typical of extended use—consistent with vessels made for a single event. A large proportion of the pottery derived from cooking and serving vessels, underscoring the culinary emphasis of the activities.

Twelve elite tombs east of the mosque were also excavated, and their stratified deposits contained abundant animal bones. The researchers considered and rejected the possibility that the fauna was merely fill redeposited during grave digging: the density of remains, particularly in Locality 4N, and the relatively limited fragmentation argued against it. Instead, the evidence points to offerings and feasting debris deposited over many generations. Excavation of Locality 4N, the only tomb with a large enough assemblage for temporal analysis, revealed distinct episodes of activity, and human remains of both sexes and all ages confirmed repeated family use of the tombs for interment. Given the exclusive nature of the elite cemetery, these appear to have been family-sponsored commemorations for the waungwana dead rather than public feasts open to all.

The methods were rigorous. Faunal remains were recovered using one-by-one-millimeter sieves, washed, and sorted at a field house before export under permit from the National Museums of Kenya. Identifications were made against reference collections at Harvard’s Museum of Comparative Zoology, Durham University’s Zooarchaeology Laboratory, and the Muséum National d’Histoire Naturelle in Paris. Fish were identified to family level using only four cranial elements—the premaxilla, maxilla, dentary, and angular-articular—while bovids were assigned to size classes following Brain’s standard framework, with sheep and goats differentiated using the criteria of Zeder and Lapham. The team also applied biometric techniques, including the log-size index, to cattle and chicken remains, preserving measurements following von den Driesch’s protocols.

The results paint a picture of feasting that was, taxonomically at least, surprisingly ordinary. Across all contexts, the main animals consumed were chickens, caprines—mostly goats, which outnumbered sheep twelve to three—and emperor fish of the family Lethrinidae. Cattle, often considered a high-status animal and a hallmark of feasting deposits elsewhere, were uncommon; caprines outnumbered cattle roughly six to one by specimen count, a ratio similar to that found in domestic refuse at the fourteenth-to-sixteenth-century site of Songo Mnara in Tanzania. Parrotfish and jackfish were also common among the fish, and nearly all identified species were reef-dwellers catchable with simple handlines rather than requiring the costly boats and tackle that elites had monopolized during the height of Swahili power. Sharks, dugongs, and sea turtles—markers of capital-intensive open-sea fishing—were present but rare, reaching far lower proportions than at Songo Mnara, where sharks constituted eleven percent of identified specimens.

Stratigraphic analysis of Locality 4N revealed a telling trend over time: the relative abundance of fish increased, and the ratio of caprine-sized to cattle-sized bovids grew as well. In the researchers’ memorable phrasing, Mtwapa’s feasts became “goatier and fishier” over the fifteenth to seventeenth centuries. This trajectory tracks the contraction of elite economic power as Portuguese incursions disrupted Indian Ocean trade networks and stone towns depopulated. Unable to fund spectacular offshore fishing ventures or procure large numbers of prestige cattle, the people who continued to gather at the mosque complex increasingly relied on humble, locally available foods. What marked these meals as special, the authors suggest, was not exotic ingredients but rather quantity, serving ware, preparation, spices, and—above all—the sacred context in which they were eaten, at the tombs of ancestors whose favorite foods in life may have been deliberately offered.

Perhaps the most startling findings were the animals forbidden under Islamic law. The assemblage contained remains of suids—including a nearly complete skeleton of a young piglet from Tomb 4K, bearing a cut mark on the scapula alongside carnivore puncture marks likely from a domestic cat—plus sea turtle bones, one plastron fragment with cut marks, and a small number of dugong specimens. Pork’s prohibition is unequivocal in Islamic scripture, making its presence in a mosque-associated context particularly striking. Yet the researchers caution against simple interpretations. Similar finds have been reported at Shanga, Unguja Ukuu, and even within a mosque at Songo Mnara. Religious identity, they argue, is constantly negotiated, and ethnographic work in the Lamu archipelago shows that Swahili communities there eat dugongs and sea turtles while maintaining other taboos. The remains might also represent non-Muslim participants, or in some cases intrusive animals—the giant pouched rat and domestic cat mandibles found together in one tomb probably belong to scavengers who died where they lived.

The study also yielded quieter surprises. A cluster of ring cowrie shells, Monetaria annulus, appeared in the earliest levels of Tomb 4N and nowhere else, likely left as offerings during commemorative visits—cowries having served as commodity money and ritual objects along the East African coast for centuries. Hundreds of pufferfish dermal spines, found in clusters suggesting only a few individuals, may likewise represent offerings rather than meals. The chicken bones proved biometrically fascinating: their size spread exceeded that of both ancient and modern Ethiopian comparative samples and could not be explained by sexual dimorphism alone, with tarsometatarsal spurs indicating nine males and twenty-three females. The researchers propose that at least two distinct populations, or breeds, of chickens were kept at the site—a possibility never before explored in Swahili zooarchaeology that they hope future ancient DNA or proteomic work will resolve.

Taken together, the Mtwapa assemblages tell a story of continuity and adaptation. Even as their town emptied and their economic reach contracted, the Watwafi returned to the Friday Mosque Complex to bury their dead, venerate ancestors, and share communal meals. The foods they ate became humbler over time, but the ritual itself endured, transforming an abandoned urban landscape into a living site of memory. Nearly forty years after Fleisher observed that the study of feasting had been surprisingly neglected for the ancient Swahili, this analysis demonstrates that the bones discarded around a mosque’s tombs can illuminate not only what people ate, but how communities preserved identity, negotiated faith, and honored their ancestors through two centuries of upheaval.