With the support of the National Science Foundation, Drs. Anthony Ranere, Ruth Dickau and an international team of researchers will conduct archaeological and paleoecological fieldwork over a two year period in the Middle Cauca River Valley of Colombia. The project brings together researchers and students from institutions in the United States, Great Britain and Colombia with expertise in archaeology, paleoethnobotany, paleoecology and geology. The research seeks to document the initial appearance of cultivation in the subsistence strategies of populations in the Middle Cauca River Valley and to identify the plant species that were either domesticated in this region or imported from elsewhere in South America or Central America. Research to date has demonstrated that species from Mexico to Brazil came under domestication early in the Holocene (ca. 9,000-10,000 years ago) and became widely dispersed over the next two or three millennia. A number of crops appear to have been domesticated in the seasonal tropical forests of northern South America, including arrowroot, sweet potato, leren, achira and possibly cocoyam, as well as a variety of squash. The Middle Cauca River Valley of Colombia is an especially promising region for examining the early cultivation, domestication and dispersal of plant species because of the numerous sites identified here with early Holocene deposits containing both microfossil remains and artifactual evidence for plant cultivation.
The project will focus on the recovery of plant remains from tools and sediments in archaeological sites already identified and tested by project personnel. Additionally, a site survey and testing program will be carried out aimed at locating sites that would extend the occupational history of the region back into the Late Pleistocene. Vegetational history will be established through an analysis of pollen and phytolith sequences paying particular attention to disturbance indicators reflecting cultivation activities. The identification of specific taxa utilized will rely primarily on analysis of starch grains and phytoliths, and secondarily on plant macrofossils and pollen. In order to confidently distinguish starch grains and phytoliths produced by domesticated species from those produced by their wild progenitors and relatives, botanical collecting of both domesticated and wild taxa will be undertaken to establish comprehensive reference collections for the study region.
The domestication of plants - and the subsequent development of agricultural economies - constitutes one of the most transformative episodes in human history. The intellectual merit of the project is the contribution it will make to our understanding of when, where and under what circumstances the critical change from gathering to cultivating plants first came about in tropical America.
The broader impact of the project includes providing students from several academic institutions with the opportunity to participate in both field and laboratory aspects of the research, thereby contributing to their training as future professionals. In addition, the comparative reference collections of starch grains and phytoliths will be made available on the web for use by other researchers studying the ancient use of plants in the American tropics. A GIS database with project generated data will also be put on the web.
The goal of this research was to document the early history of cultivation and plant domestication in the Middle Cauca River Valley of Colombia and to examine the timing of the introduction into the region of plants domesticated elsewhere in the American tropics. The location of this region in the northwestern corner of South America is a good place to monitor the dispersal of domesticated plants moving north from other regions in South America and south from Mexico and Central America. Importantly, the Middle Cauca region contained a number of archaeological sites that were occupied during the Early and Middle Holocene (ca 12,500 to 5000 cal years BP) when early cultivation, plant domestication and dispersals occurred. We planned to focus on the recovery of starch grains, phytoliths, pollen and macrobotanical remains in order to identify the plants being utilized through time. That required that we have access to sediment samples and stone tools in contexts confidently dated to the Early and Middle Holocene from sites within our study region. To this end we obtained 36 AMS radiocarbon dates from 11 different sites, bringing the total number of radiocarbon dates for the sites under investigation to 65. The five sites with the longest history of occupation all have between 8 and 15 radiocarbon dates to firmly establish the chronological sequence in each. The dates on preceramic cultural deposits range from 12,574 + 124 years (calibrated) to 4107 + 85 years (calibrated). Starch grains are preserved on the surfaces of stone tools used to prepare plant foods, but not in sediments. Phytoliths (silicified plant cells) are found both on stone tools and in sediments. Pollen is primarily recovered in sediments although some pollen grains are recovered from stone tool surfaces. Over the course of two field seasons we conducted major excavations at La Mikela, Cuba and Invias 3. In addition to these three sites and 6 other preceramic sites investigated by members of our research team, we also have samples from 8 additional sites recently excavated by colleagues from the region. The single occupation that dates back to 12,574 + 124 cal BP is from the Cuba site, located in the Hacienda Cuba coffee plantation within the city limits of Pereira. The single handstone fragment yielded nine starch grains of unknown taxa. However, none of the grains belonged to any of the well know tropical domesticates nor any of their close relatives. We tentatively interpret this to mean that the plants represented on this handstone were wild species that were gathered and/or possibly cultivated. The Cuba site has yielded starch grains dating from 11,255 + 46 cal BP of achira (Canna sp., probably indica) a root crop that was likely domesticated in northern South America. Starch from cocoyam (Xanthosoma sp., probably sagittifolium) was recovered from the Invias 3 site in a level dated to 10,255 + 99 cal BP. Because this genus has a number of species that occur naturally in this region, we cannot say definitively that the starch grains are from the domesticated species but this seems probable. By 8000 cal BP, a number of domesticated species in addition to achira and cocoyam were recorded in the starch grain record for the region. Maize (Zea mays) and manioc (Manihot esculenta) had entered the region from their distant origins in Mexico and the Brazil-Bolivia borderlands. Domesticated species from homelands somewhere in northern South America, perhaps even the Cauca Valley, include arrowroot (Maranta arundinaceae), leren (Calathea sp.), and yam (Dioscorea sp.). Throughout the sequence the stone tool assemblages remained remarkably similar. The most characteristic tools are handstones of unmodified river cobbles and boulder bases used for mashing and grinding plant foods. Also occurring throughout the sequence are large bifacially flaked and normally notched hoes/spades made of medium to coarse grained volcanics. The earliest examples have chipped stone bits while later examples have ground and polished bits. They were apparently used for harvesting roots and probably for planting crops as well. The rest of the chipped stone industry is characterized by the haphazard removal of flakes from irregular cores that are then selected for use with little or no purposeful retouch. Most of the primary material is locally available medium to coarse grained volcanics (e.g., andesite, dacite, basalt) and quartz. Small numbers of irregular cores as well as cobble hammerstones are also recovered throughout the preceramic sequence. The project was a collaborative effort by archaeologists, paleoethnobotanists, palynologists, geologists and soil scientists from the United States, England and Colombia. It provided training for graduate and undergraduate students from Temple University, Universidad Tecnologica de Pereira, Universidad de Antioquia and Universidad de los Andes. The information provided by this research will contribute to our understanding of the timing and nature of plant domestication and dispersals in the American tropics.
