Peanuts are a major staple in the diets of many peoples of the world. Peanuts are also an important commodity for many economies. The peanut's original home is believed to be the slopes of the Andes in Brazil and in Peru. Portuguese traders, explorers, and missionaries transported the peanut to Africa and Spain. From Africa peanuts traveled by ship to North America and were grown on farms in the southern British colonies.

Today's domesticated hybrid peanuts are different from the original wild groundnuts of South America. These wild groundnuts are able to survive in more rugged conditions. In search of the peanut's origins, scientists hope to locate the hardy B-genome parents. They believe that enhancing today's peanut varieties with the B-genome of its ancestor would benefit farmers and consumers around the world, especially the poor. Today, there are 68 known wild peanut species.

Researcher David Williams is a plant explorer and ethnobotanist who is based in Cali, Colombia, at the Americas Office of the International Plant Genetic Resources Institute, a Future Harvest Center of the Consultative Group on International Agricultural Research (CGIAR). Williams says, "The species we are looking for could eliminate much of the need for farmers to use pesticides and also help them cope with drought." Williams is using a GIS application called FloraMap to map the location, species, and habitat of various species in the regions of northwestern Paraguay, southeastern Bolivia, and southeastern Brazil.

FloraMap was developed by the International Center for Tropical Agriculture (CIAT), another Future Harvest Center of CGIAR, and links to agroclimatic and other databases. It allows biodiversity specialists to create maps showing the most likely distribution of wild species in nature. FloraMap was built using Esri's MapObjects LT software. It is an application that helps map the distribution of wild organisms based on knowledge of the sites from which they have been collected. It works from an interpolated climate surface and makes a probability model of the climates of the collecting sites. Each pixel of the climate surface is then compared with this model to determine the probability that it could come from the original population.

The application is programmed in Borland C++ Builder. MapObjects LT is used to display Esri shapefiles as background for the maps and then to overlay a rendered shapefile of the probability surface that was constructed in C++. Pan and zoom functions are implemented as are other functions, such as defining subsets of the collection points, using a cursor polygon draw function and distance calculation.

Williams and CIAT researcher Andrew Jarvis recently mapped the coordinates where the B-genome and other wild peanut species are most likely to be found. "We combined meteorological station data and a large global climate change model called Hadley with the geographic coordinates of the sites where wild species had been collected in the past," says Jarvis. "From there, we used FloraMap to map and compare distribution patterns, both present and future, for each target species."

For more information, contact Peter Jones (e-mail: p.jones@cgiar.org), or visit the FloraMap Web site (www.floramap-ciat.org).