Archéozoologie et Archéobotanique

The identification of archaeological fruits and seeds is crucial for understanding the relationships between humans and plants within the cultural and biological history of both wild and cultivated species. We compared the relative performance of a deep learning approach, namely convolutional neural networks (CNN), and outline analyses via geometric morphometrics using elliptical Fourier transforms (EFT) at identifying pairs of plant taxa. We used their seeds and fruit stones that are the most abundant organs in archaeobotanical assemblages, and whose morphological identification, chiefly between wild and domesticated types, allows to document their domestication and biogeographical history. We used existing modern datasets of four plant taxa (barley, olive, date palm and grapevine) corresponding to photographs of two orthogonal views of their seeds that were analysed separately to offer a larger spectrum of shape diversity. Sample sizes ranged from 473 to 1,769 seeds per class, which constitute a relatively small dataset for training CNNs models yet typical within archaeobotanical research. On these eight datasets, we compared the performance of CNN and EFT coupled with linear discriminant analyses. Our objectives were twofold: i) to test whether CNN can beat geometric morphometrics in taxonomic identification and if so, ii) to test which minimal sample size is required. We ran simulations on the full datasets and also on subsets, starting from 50 images in each binary class. For the CNN network, we deliberately used a candid approach relying on pre-parameterised VGG19 network. For EFT, we used a state-of-the art morphometrical pipeline. The main difference rests in the data used by each model: our CNN used bare photographs where EFT used outline coordinates. This “pre-distilled” geometrical description of seed outlines is often the most time-consuming part of morphometric studies. Results show that our CNN beats EFT in most cases, even for very small datasets. We finally discuss the potential of CNNs for archaeobotany, and how bioarchaeological studies could embrace both approaches, used in a complementary way, to better assess and understand the past history of species.

The Muse oil-shale beds (Muse OSB) of the Permian Muse Formation in the Autun Basin (France) have received attention for their exceptionally well-preserved flora and fauna throughout many decades. Recent CA-ID-TIMS U/Pb radiometric datings place the Muse section in the lower Asselian (~298.6 Ma), leading to a re-examination of the palynoflora using an alternative palynological method. Twelve productive samples provide a detailed taxonomic and quantitative evaluation of the palynoflora. The new data revealed now two distinct palynological assemblages, Muse-A in the lower part and Muse-B in the middle-upper part of the Muse section, assigned to a peat swamp forest environment dominated by cordaitalean conifers, alongside medullosalean cycads and callistophytalean seed ferns, with an undergrowth of lycophytes and ferns. Additionally, a forest dominated by walchian conifers (Voltziales) would be present but more distant from the lake. Vegetation dynamics between these associations are generally stable, except for an increase in Cordaitales at the expense of Voltziales. This variation, observed throughout the Autun Basin, seems to be affected by seasonal climate-driven changes. The co-occurrence in both assemblages of Permian xerophytic elements accompanied by Carboniferous wetland flora suggests a seasonally dry climate, supporting hygrophyte plant survival, persisting for at least 300 ky after the Carboniferous–Permian boundary. The phytoplankton community consisted of freshwater green algae, increasing towards the upper part of the section, coeval with increased explosive volcanic activity. Wind-transported volcanic ash likely provided nutrient input, enhancing lacustrine phytoplankton growth.

• Archaeobotany successfully records viticulture in the absence of specialised production facilities. • Similar trends in seeds and charcoals suggest local viticulture by the 6th-5th c. BCE. • Winegrowing has developed in a favourable climatic context. • Taphonomic causes prevent archaeobotany from accurately recording the scale of specialised viticulture. Winegrowing has dramatically shaped the landscapes as well as the culture and economy of societies in the Mediterranean region. It is generally accepted that it appeared and expanded in Mediterranean France during the Iron Age and the Roman period (ca 750 BCE − 500 CE). Viticulture flourished massively during the Early Roman Empire, when wine was widely exported throughout the Empire. The objective of this paper is to propose an updated overview of the history of viticulture based on a multidisciplinary survey combining archaeobotanical, archaeological data and modelling of wine potential yields. We aim to compare the information provided by seed-fruits and by charcoal, and to better understand how these data can be used to trace ancient viticulture in relation to other archaeological data. We collected and entered into a database all available results of charcoal and seed-fruit analyses from Mediterranean France and the middle Rhône valley. Seed and fruits provide evidence of fruit consumption and processing (table vs . pressing), while charcoal remains are stronger indicators of local cultivation. We also included published archaeological information about wineries and amphorae factories. This provides another way of identifying areas of cultivation and processing into wine, and gives hints on the scale of production and the possible destination for export. The reconstructions based on the multi-proxy information contained in the database are compared to the climate-vegetation LPJmL model estimates of wine potential yields and their variations over space and time. Charcoal and seed evidence concur to show that viticulture started in the 6th-5th c. BCE and developed with the progressive increase of potential wine yields throughout the Iron Age and then with the Roman Climatic Optimum. In the Roman period, the development of specialised winemaking facilities provides impressive evidence of viticulture, but is at the same time less favourable to its record by archaeobotany. Apparently, charcoals and seeds are more frequently preserved in rubbish assemblages when winemaking activities are still associated with domestic activities. While wineries and amphorae factories bear witness to a wine crisis in the Late Roman period, archaeobotanical remains indicate that viticulture persisted, probably in a less specialised mode, and modelling suggests that potential wine yields did not decline in comparison with the heyday of regional viticulture. Archaeobotany also provides consistent evidence of the existence of grapevine cultivation in urban Roman contexts.

The identification of archaeological cetacean specimens to the species level often presents challenges, primarily due to the frequent fragmentation of their remains. This limitation hampers our ability to comprehensively understand the spatiotemporal distributions of cetaceans in the past and constrains our knowledge of early whaling activities. To address this issue, a systematic zooarchaeological review was conducted, encompassing published literature and reports that compile available data related to cetaceans retrieved from archaeological contexts in Portugal, spanning from the Middle Palaeolithic to the 18th century. Furthermore, this study introduces a new discovery of a cetacean specimen from the Roman Republican levels at Castelo Velho de Safara, which has been incorporated into the aforementioned dataset. The taxonomic identification of this specimen was accomplished using Zooarchaeology by Mass-Spectrometry (ZooMS). The study confirms that cetacean remains have been present in Portuguese waters since the Middle Palaeolithic, with higher concentrations in the southern and central regions of the country. ZooMS analysis confirmed the presence of the North Atlantic right whale at the inland site of Castelo Velho de Safara, which supports whale product trading during the Roman Republican period in Portugal. Evidence of cetacean exploitation continues to the Medieval and Modern periods, reflecting a strong cultural link between Portuguese ancient culture and whaling practices.

The cultivation of grapevines has spanned millennia, leading to thousands of varieties through exchanges, mutations, and crosses between genotypes, as well probably as gene flow from wild populations. These varieties are typically categorized by regional origin and primary use, either for wine production or fruit consumption. France, within the Western European group, hosts many of the world’s renowned wine grape varieties. However, the historical development of cultivated grapevines in France and in the world remains poorly understood. This study applies morphometry on 19,377 charred and waterlogged archaeological grape pips to investigate the evolutionary history of grapevine in France over the last 10,000 years. The study compares seed outlines and lengths, corrected for taphonomic distortions, with a reference collection of 80 wild and 466 modern domestic grapevine accessions. Findings reveal a shift from wild grapevine exploitation to the expansion of domestic varieties around 600–500 BCE, coinciding with Mediterranean cultural influences and the introduction of eastern grape types. The identification of the East-Table group, a group of varieties of eastern origin for fruit consumption, indicates that grapes were also grown for food, especially in Mediterranean regions and near urban areas, alongside wine production. Early French viticulture featured a notable presence of Western European wine-type grapevines. The abundance of pips with wild-like morphology suggests early cultivation involved plants at an initial domestication stage and gene flow between introduced and wild grapevines. As viticulture spread northward, wild and Eastern morphotypes declined, leading to the dominance of Western European wine types in inner France during the Middle Ages.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-024-72692-6.

Pottery vessels often comprise major burial goods at archaeological sites, thus providing valuable information for reconstructing past mortuary practices. However, because of the uncertainty of its function or use, which has been interpreted mostly through typological studies alone, the analytical potential of pottery as a burial good has not been fully exploited. This study applied bio-chemical and geochemical analyses for the first time to funerary pottery vessels of the Iron Age of North Iran to examine their function and use. The study materials are from the necropolis of Ghalekuti, Dailaman, excavated in the 1960s. Direct radiocarbon dating conducted on human and animal bones in the graves and typological analysis of the pottery anchored the chronological position of the pottery materials to the 2nd and 1st millennium BC. A petrographic analysis revealed that pottery vessels can be classified into six fabric types, including those with coarse tempers that are effective for cooking. Pottery pastes with finer inclusions less suited for cooking appeared during the early first millennium BC (Iron Age III). To obtain further insight into the function of the pottery, we conducted organic residue analyses. The results demonstrated that the vessels retained remains of botanical and animal origin. In particular, jars with tubular spouts, characteristic of the Iron Age III period, were likely specialised for botanical products. Interestingly, both carcass and dairy products from ruminant animals (cattle and caprine) were processed in short-neck jars and bowls, including spouted bowls, suggesting their use in a liquid state. Products from ruminants, particularly dairy products, may have played a significant role in the daily and ritual use of pottery vessels during the study period in Northern Iran. These results indicate that a range of pottery vessels used for specific purposes before the burial was offered for graves, helping us better understand the mortuary practices of Iron Age Iran.

Societal Impact StatementAgrobiodiversity is central to sustainable farming worldwide. Cultivation, conservation and reintroduction of diverse plant species, including ‘forgotten’ and ‘underutilized’ crops, contribute to global agrobiodiversity, living ecosystems and sustainable food production. Such efforts benefit from traditional and historical knowledge of crop plants' evolutionary and cultural trajectories. This review is a first attempt at systematically gauging species representativeness in studies of archaeological plant remains. Results indicate that, in addition to discipline‐specific methodological sources of bias, modern agricultural biases may replicate themselves in crop history research and influence understandings of ‘forgotten crops’. Recognizing these biases is an initial stride towards rectifying them and promoting agrobiodiversity in both research and practical applications.SummarySo‐called ‘forgotten’ or ‘orphan’ crops are an important component of strategies aimed at preserving and promoting biodiversity. Knowledge of historical cultivation, usage, and geographic and evolutionary trajectories of plants, that is, crop history research, is important for the long‐term success of such efforts. However, research biases in the crops chosen for study may present hurdles. This review attempts to systematically identify patterns in crop species representativeness within archaeology‐based crop history research. A meta‐analysis and synthesis of archaeobotanical evidence (and lack thereof) is presented for 268 species known to have been cultivated for food prior to 1492 CE from the Mediterranean region to South Asia. We identified 39 genera with known crop plants in this geographical and historical context that are currently absent from its archaeobotanical record, constituting ‘orphan’ crops of archaeobotany. In addition, a worldwide synthesis of crop species studied using geometric morphometric, archaeogenetic and stable isotope analyses of archaeological plant remains is presented, and biases in the species represented in these disciplines are discussed. Both disciplinary methodological biases and economic agenda‐based biases affecting species representativeness in crop history research are apparent. This study also highlights the limited geographic diffusion of most crops and the potential for deeper historical perspectives on how crops become marginalized and ‘forgotten’.