
For the first time, a study has identified an unmistakable “chemical signature” of intensive maize consumption among precolonial Brazilian populations. The isotopic signal, based on naturally occurring variants of carbon, nitrogen, and oxygen, was found in human bones and teeth dated to between 1,055 and 333 years before present from archaeological sites in the Brazilian Cerrado—a region that has historically been marginalized from ecological and anthropological perspectives compared with areas such as the Amazon.
Although the level of consumption identified is comparable to that of Maya or Andean populations, the study published in Science Advances indicates that the maize-based polyculture system was integrated with complex landscape management practices, which included other plant and animal resources.
For André Strauss, professor at USP’s Museum of Archaeology and Ethnology (MAE) and corresponding author of the article, finding this evidence in the Cerrado is something of an irony of fate. “It is a drier place, with less rainfall, and Cerrado societies did not produce particularly elaborate material culture. That is one of the reasons why the biome has always been cast in a secondary role. Yet it is precisely there that this evidence of maize consumption appears in Brazil, among these Aratu societies, which produced this highly distinctive Cerrado pottery and inhabited large circular villages that have no parallel in the ethnographic record”, he told the Jornal da USP.
The findings reinforce the advanced agricultural knowledge of these populations, since growing maize in this environment required mastery of irrigation and crop rotation techniques adapted to the local ecosystem conditions.
Eliane Chim, the publication’s first author and a doctoral student at the MAE, said the study helps challenge the idea that these populations lived under constant resource scarcity. “People were selecting, planting, and working with this diversity of maize. We know, in particular, that the Cerrado has a wide diversity of landrace seeds that require highly sophisticated land-use techniques. Separate fields are needed so that one variety does not cross-pollinate with another”, she explained. According to the researcher, this knowledge was maintained over generations.
The researchers also challenged another recurring interpretation in Brazilian archaeology: that Tupi peoples did not occupy the Cerrado because they considered it an inhospitable environment. “We see some studies saying that they were not interested in the Cerrado because it was not forested or because it was sparsely populated. But what we see is the opposite: many people lived there. It was not scarcity that kept them away; perhaps it was because the territory was already inhabited”, said Eliane Chim.
“This discovery makes it possible to directly study the role of maize in precolonial societies after it arrived in different parts of South America”, Patrick Roberts, Director of the Department of Land Use and Urbanization Coevolution at the Max Planck Institute of Geoanthropology, told the Jornal da USP. “Our data allow us to investigate how these adaptations occurred across the diverse environments of the Cerrado, while much of the attention is usually focused on the Amazon Basin and its tropical forest biomes”, emphasized Roberts, who is also a co-author of the article.
Biologist and former president of the Coordination for the Improvement of Higher Education Personnel (Capes), Mercedes Bustamante, one of the leading experts on the Cerrado biome, did not participate in the study but commented that the research fills knowledge gaps with robust evidence. At the same time, it prompts reflection on current land-use models in the Cerrado. “Allowing for differences in scale and time frame, it is thought-provoking to consider the possibility of alternative and diversified agrifood systems, already advocated in scientific studies and traditional knowledge, capable of ensuring food, water, and climate security while conserving biodiversity”, said the professor at the University of Brasília (UnB).


Dietary patterns of Cerrado peoples
To understand the subsistence strategies of these societies, the researchers used stable isotope analysis—a tool capable of reconstructing diets and the geographic origin of foods. The team analyzed carbon, nitrogen, and oxygen isotope signatures found in the teeth and bones of 101 individuals recovered from 37 archaeological sites in the Cerrado, Caatinga, and Atlantic Forest biomes. In addition to the human remains, samples of local fauna were also examined, as they were essential for establishing comparative parameters capable of distinguishing the consumption of different food resources.
The analyses concern Late Holocene populations, a period encompassing the last 4,200 years of geological history, associated with the Aratu ceramic tradition—societies that lived in large open-air villages—and the Una tradition—producers of small, simple pottery with rounded bases, generally found in rock shelters.
According to the study, individuals associated with the Aratu tradition showed significant enrichment in carbon-13, a signature known as “C4”, typical of plants such as maize and sugarcane. Since sugarcane did not exist in precolonial Brazil, an isotopic signal from C4 plants is understood as evidence of maize consumption.
The populations that occupied rock shelters, by contrast, had significantly more diversified diets, combining foods derived from C3 plants, C4 plants, and animal resources. “C3 plants include almost all existing plants. They include trees and virtually all the foods you eat, such as potatoes, cassava, watermelon, and all kinds of fruit”, said Eliane Chim.
The chemical data analyzed do not make it possible to determine precisely which species these people consumed. However, botanical evidence found at the archaeological sites themselves provides a good indication of crop diversity. “We find beans, peanuts, cassava, and squash”, Strauss recalled. “There were also fruits… cashews, guavas, and passion fruit”, the archaeologist added.



The evidence also points to modest meat consumption, with variations among individuals. The researchers emphasize, however, that this should not necessarily be interpreted as limited access to animals, but rather as a reflection of food choices or specific cultural patterns.
“There was this remarkable cultural diversity in the Cerrado, even though the villages were located very close to the shelters. These people coexisted, lived during the same period, and we have skeletons dating from the same time at sites within a 50-kilometer radius” — Eliane Chim
The results also shed light on another debate in Brazilian archaeology: whether Una and Aratu represented the same society using different types of settlements. Under this hypothesis, the group would have lived in large open-air villages and buried their dead in large funerary urns, while occasionally visiting rock shelters, where they carried out burials and other activities. “The isotope data show us that the diet of the individuals buried in the shelters was completely different from that of the individuals found in the villages, strengthening the hypothesis that they were also culturally distinct”, Eliane Chim pointed out.
“Fundamentally, the Aratu represent a striking shift in the archaeological pattern of Central Brazil. Until then, there were the Lagoa Santa hunter-gatherers, followed by pottery-producing peoples who made very small, finely crafted vessels, known as Una pottery. Then, suddenly, the Aratu appear, bringing three innovations all at once: the large Aratu urns, the large circular villages, and—as we now know for certain—the cultivation of maize, which formed the basis of their diet”, André Strauss emphasized.
“The study sheds light on the cultural diversity of the Cerrado and on how this diversity interacted with biological diversity, revealing a rich socioecological system that deserves recognition and prominence”, Mercedes Bustamante stated.

Uncovering the archaeology of the Americas
The extensive sample set presented in the Science Advances publication is part of Eliane Chim’s doctoral research at USP’s Museum of Archaeology and Ethnology (MAE), under the supervision of André Strauss. Her dissertation is expected to be presented as a collection of scientific articles, this being the first in the series.
In addition to the diets of Late Holocene populations, the research examines skeletons dating from 10,000 to 8,000 years ago, from the Early Holocene, and from 6,000 to 4,000 years ago, from the Middle Holocene. The challenge was to cover the dispersal of these peoples across the territory. For this reason, the study brought together 31 researchers from ten Brazilian universities, in collaboration with the Max Planck Institute in Germany and institutions in Portugal, Poland, and the United States.
The work represents a milestone for Brazilian archaeology, as it is the first study in the field from Brazil to be published in a high-impact journal with both the first and last authors being Brazilian.
“We established this collaboration with several institutions in Brazil that hold collections in their museums and archaeology laboratories. Archaeologists have been excavating since the 1970s to assemble this large set of individuals”, said Eliane Chim.
In addition to collecting samples and accessing Brazilian collections, the researcher was able to refine her analysis of stable isotopes in human and animal remains at the Geoanthropology laboratory of the Max Planck Institute in Germany through an exchange program. “Although I had already been collaborating with USP researchers, this type of exchange strengthens partnerships and gives them lasting significance by promoting the training and development of early-career Brazilian researchers”, said Patrick Roberts.
According to Strauss, the study stood out for increasing the number of radiocarbon dates for Aratu occupations by approximately 50%. “Knowing when they began and ended, and when they reached a new location, is also very important. This was a relatively poorly characterized area of Brazilian archaeology”, he said. “They were everywhere. It was a society that was truly very well adapted, with great demographic success and a degree of cultural standardization that is also striking. To me, it is the most beautiful pottery in Brazil”, he added.
For Eliane Chim, the most surprising finding was the longevity of the culture: “We thought the Aratu had disappeared 600 years ago because of conflicts and the arrival of Tupi peoples along the edges of the Cerrado. But they continued until the Portuguese invasion. We have a site in the Cerrado dated to 400 years ago.”
The article Maize-Based Polyculture, Not Monoculture, Sustained Precolonial Societies in the Brazilian Cerrado was published in Science Advances and is available online.
For further information: Eliane Chim, chim@usp.br; André Strauss, strauss@usp.br
*Written by Tabita Said
English version: Nexus Traduções, edited by Denis Pacheco


