Latitudinal gradient in dairy production with the introduction of farming in Atlantic Europe

The introduction of farming in Europe significantly impacted health, social structures, and demography. While the spread of domesticated plants and animals is well-documented, the development of these early economies within diverse environmental and cultural settings remains unclear. Traditional explanations have focused on demic diffusion versus acculturation, but more recent research emphasizes regional variations and complex models considering geographic and climatic factors, as well as interactions between pioneer farmers and indigenous hunter-gatherer-fisher populations. Genetic studies support a Near Eastern origin for Europe's earliest farming communities, but provide limited detail on the social or economic drivers of the Neolithic. Scientific methods, such as stable isotope analysis of human remains and organic residue analysis of pottery, are revealing the varied nature of early farming communities. Dairy production, previously believed to be a later development, was present in early Neolithic economies and may have motivated ruminant domestication. However, the scale of dairying relative to meat production remains unknown, with studies highlighting regional variations in early animal husbandry. The Atlantic coast of Europe has received limited attention, with prior studies primarily focusing on Britain and Ireland. This region had a high density of Late Mesolithic sites, where fish and shellfish were heavily exploited. The transition to farming along these coastlines and estuaries, and the replacement of hunter-gatherer-fisher subsistence practices, are debated. One might expect continuity in the use of marine resources in early Neolithic pottery, particularly near previous hunter-gatherer-fisher activity. However, paleodietary reconstructions using stable isotope analysis of skeletal remains often refute this, showing little evidence for continued aquatic protein consumption with the onset of the Neolithic. The degree of farmer-forager interaction is also debated, with some evidence of limited admixture between the two groups. This research aims to address these gaps by examining organic residues from pottery across a wide latitudinal range of the Atlantic coast of Europe, exploring regional economic variations and interactions with hunter-gatherer-fisher communities. The study incorporates new data and previously published datasets to examine pottery use and animal exploitation by early farmers across various regions with different Mesolithic site densities.

Prior research on the introduction of farming in Europe has focused on polarized hypotheses such as demic diffusion versus acculturation. More recently, the importance of regional variations and complex interactions between farmers and foragers has been emphasized. Genetic analyses have largely supported a Near Eastern origin for early European farming communities. However, these studies offer limited insights into the economic and social factors that drove the Neolithic transition. Studies utilizing stable isotope analysis of human remains and organic residue analysis of pottery have provided valuable insights into the diverse nature of early farming communities and the importance of dairying in even the earliest Neolithic economies. While evidence for dairy production exists in various regions of Europe, the scale and intensity of dairying relative to meat production remain uncertain. Research on the Atlantic coast of Europe has been limited, particularly compared to other regions. The existing research emphasizes the significance of fish and shellfish in the Late Mesolithic period. The interplay between the introduction of farming and the continuation of fishing and shellfish exploitation is a matter of ongoing discussion. The interaction between Mesolithic foragers and Neolithic farmers is debated and genomic analysis points to some admixture but its timing and location are not fully understood. Studies have suggested variations in early agro-pastoral economies along the Atlantic margin but regional differences remain understudied.

This study analyzed lipid residues from 246 pottery sherds from 24 Early Neolithic sites spanning from Portugal to Normandy and the Western Baltic. Pottery from the initial phases of the Neolithic expansion was selected from small and fragmented assemblages. The data were combined with previously published data from 39 archaeological sites to create a supra-regional perspective on pottery use and animal exploitation from ca. 5500–3500 cal BC. The samples were analyzed using well-established procedures for lipid extraction, including acidified methanol and solvent extraction. Gas chromatography-flame ionization detection (GC-FID), gas chromatography-mass spectrometry (GC-MS), and gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS) were employed to identify and quantify lipids. Stable carbon isotope (δ13C) values of C16:0 and C18:0 fatty acids were measured to distinguish the source of animal fats (ruminant adipose, ruminant dairy, and porcine). A Bayesian mixing model was used to account for the mixing of different foodstuffs and to estimate the proportions of lipids derived from various animal products in each region. The AverageR and SpreadR models were employed to spatially interpolate Δ13C values and estimate the earliest arrival of the Neolithic, respectively. These models provided spatial estimates of Δ13C values and uncertainties expressed as standard errors of the mean. Specifically, the AverageR model utilized a generalized additive mixed model with a thin plate regression spline to incorporate both intra-site and inter-site variation. The SpreadR model used an extreme quantile approach (0.5% quantile) estimated by Bayesian quantile regression with a spline smoother. Both models provided standard errors of the mean as a measure of uncertainty. Detailed descriptions of the sample information, lipid extraction and analytical protocols, GC-FID, GC-MS, GC-C-IRMS, data analysis, mixing model, and spatial interpolation are provided in the supplementary materials.

Analysis of 246 pottery sherds revealed interpretable lipids in 95% of the samples. Lipid preservation was excellent in samples from the British Isles, Ireland, and Northern Europe, unlike samples from the Mediterranean and Southeastern Europe. The distribution of fatty acids in the Iberian Peninsula and French sites was consistent with degraded animal fats. The stable carbon isotope (δ13C) values of C16:0 and C18:0 were used to distinguish the source of animal fats, including ruminant adipose, ruminant dairy, and porcine fats. A significant latitudinal gradient was observed in the Δ13C values, with a greater proportion of pottery sherds from higher latitudes containing dairy products. Stable carbon isotope values typical of marine oils were only found in pottery from the Western Baltic, consistent with the presence of aquatic lipid biomarkers in those vessels. This contrasts with the absence of aquatic biomarkers in pottery from other Atlantic sites, despite sensitive protocols for their detection. A Bayesian mixing model confirmed the increasing proportion of dairy lipids compared with meat lipids as latitude increased, excluding the Western Baltic where the data showed a bimodal distribution. The proportion of dairy and ruminant carcass fats varied considerably across the regions. Faunal data from the Iberian Peninsula show ovicaprines dominating assemblages, while cattle were more important in Northern France, UK and Denmark. This indicates a connection between intensive dairying and cattle-based economies, with sheep and goats exploited for meat and milk in the early phases of the Neolithic. The frequency of dairy residues in pottery correlated with the date of the introduction of domesticates to the area. In Central and Southern Portugal, farming arrived with ovicaprine-based economies and the pottery showed a mixed meat/milk economy. In Northern Spain, the increased frequency of dairy residues may reflect secondary Neolithisation. In Atlantic France, pottery showed both ruminant milk and meat processing, consistent with Early Neolithic cattle remains from Northern France and Central Europe. In Britain and Ireland, over 80% of Early Neolithic pottery contained dairy fats, more frequent at higher-latitude sites. The analysis also revealed a clear absence of marine resources (except the Western Baltic) in early Neolithic pottery from the Atlantic coast, suggesting a significant shift in dietary practices with the introduction of agriculture. The Western Baltic, which had a tradition of pottery use from the Mesolithic hunter-gatherer-fishers, shows a different scenario where both marine and terrestrial resources were present in the pottery.

The findings demonstrate significant geographic differences in pottery use and animal management along the Atlantic coast of Europe. The increased proportion of dairy residues in pottery from higher latitudes suggests that intensive dairying was linked to cattle-based economies. The correlation between dairy residues and latitude, even considering the temporal span of the Neolithic dispersal, highlights the importance of local environmental factors and cultural adaptations. The study shows how regional responses to the arrival of farming were influenced by farmer traditions, local environments, and forager responses. Economic adaptations were necessary before higher latitudes could be fully utilized for food production, leading to hiatuses in the Neolithic expansion. The absence of marine resources in most sites contrasts with the Western Baltic, where continuous pottery use from the Mesolithic facilitated the continued exploitation of aquatic resources. The variation in the scale of dairying across the latitudinal transect, despite early populations being largely lactose intolerant, suggests that this practice may have influenced the selection pressure for adult lactase persistence, a hypothesis supported by the high selection pressure for this trait in Northwest Europe.

This study reveals a latitudinal gradient in dairy production across the Atlantic coast of Europe during the early Neolithic, highlighting the adaptation and economic strategies of early farming communities. The absence of marine resources in most Early Neolithic pottery indicates a significant shift in subsistence practices. The correlation between dairy production and latitude suggests a complex interplay between environmental factors and cultural adaptations. Further research could investigate the specific environmental constraints and cultural factors that shaped the development of dairy practices in these regions, and its influence on the evolution of lactase persistence.

The study's conclusions are based on organic residue analysis of pottery, which may not fully capture the complete dietary practices of the populations. The preservation of lipids in pottery can be affected by various factors, potentially introducing bias. While the study uses a large dataset, the distribution of sites may not be perfectly uniform across the region, potentially influencing the interpretation of spatial patterns. Future research could benefit from more extensive and more evenly distributed sites across the studied region. The study also focuses solely on pottery residues and does not directly address other aspects of early Neolithic economies.