About maize, manioc and agricultural production in pre-Columbian Amazonia

A new paper from Dickau et al., recently published on-line by the Journal of Archaeological Sciences, brings us back to one of the favourite themes of this blog: pre-Columbian agriculture in the Amazon Basin. The work of Dickau et al. confirms the findings of Bruno (2010), also a co-author in Dickau et al., and provides more data from new sites. They have analysed botanical remains from 2 pre-Columbian monumental mounds east of Trinidad (Loma Salvatierra and Loma Mendoza) and from another site, a ring ditched village called “Granja del padre”, about 200 Km far from the mounds, close to Bella Vista(see location in Fig 1 and photo of the ring village in Fig 1a). Analysis of macro and micro botanical remains from Loma Salvatierra and Mendoza and from Granja del padre suggests that the most common cultivated species was maize (Zea mais L.). It was present in almost all samples and fairly abundant. The second most common cultivated species seems to have been manioc  (Manihot esculenta Crantz). Interestingly, maize was also more frequently encountered on ceramic graters, which were thought to be processing tools for grating manioc . They recovered starch grains from artefacts and clearly identified 5 grain of yucca vs. 115 of maize.  This is really surprising because manioc is a good source of energy and far easier to cultivate than maize. Ceramic graters  can be found all over the Llanos. It can be easily assumed that manioc was cultivated all over the Llanos de Moxos because, provided good drainage, manioc can grow even on very bad, acid, aluminium rich soils. I was actually persuaded that raised fields, elevated earth platforms that were user as agricultural surface, were built to provide drained land for manioc. But now, it seems that maize, and not manioc, was the most important crop in the Llanos de Moxos.

Figure 1. The Llanos de Moxos. Red triangles are the Salvatierra and Mendoza mounds. a) Google Earth photo of the Granja del padre with the approximate locations of the archaeological excavations in yellow.

We have to consider that starch grain analysis is still a developing technique and the variables that affect preservation of different taxa are not yet fully understood. But, if this data is confirmed by other researches, we could start considering ceramic graters (Fig 2) as a proxy for maize consumption. As ceramic graters are found everywhere in the Llanos de Moxos, a direct link between ceramic grates and maize would indicate that maize was used all over the region.

Figure 2. Fragment of a ceramic grater from the Llanos de Moxos

This opens up a very interesting topic for future research: was maize cultivated in all the sites where it was consumed? This is not a trivial question because maize is a very demanding crop: it needs many nutrients and a lot of water, but rots if the soil is waterlogged. In the Llanos de Moxos soils and hydrology change a lot from one place to another. For example, while soils in the region of the Salvatierra and Mendoza mounds form over fertile mid-Holocene fluvial sediments (Lombardo et al. coming soon, I hope J ), la Granja del padre is found on saprolites (rotten granite rocks), very acid and poor in nutrients. Moreover, in the region of Bella Vista, no raised fields or Amazonian Dark Earths have been reported. So, did they grow maize using slash and burn agriculture? If so, this would have greatly limited potential population density in the area, as slash and burn agriculture is considerably extensive. Each family needs at least 30-40 hectares of forest for cultivation and far more as a reservoir for hunting and medicinal plants. Or was maize “imported” from other regions? Exchanged for other goods? The estimation of pre-Columbian population density, and the extent of pre-Columbian de-forestation, is a controversial issues among ecologists, geographers and archaeologists working in Amazonia. More studies like that of Dickau et al. are needed in order to shed some light on the past of Amazonia during the late Holocene. Their work adds important new data to the discussion, although we are still far from fully understanding what was going on in the Bolivian lowlands between 2000 and 500 BP.

Ruth Dickau, Maria C. Bruno, José Iriarte, Heiko Prümers, Carla Jaimes Betancourt Irene Holst, & Francis E. Mayle (2011). Diversity of cultivars and other plant resources used at habitation sites in the Llanos de Mojos, Beni, Bolivia: Evidence from macrobotanical remains, starch grains, and phytoliths Journal of Archaeological Science : 10.1016/j.jas.2011.09.021