Public Release: 

Mexican farmers effectively cultivate phenotypic diversity in maize

Institut de recherche pour le développement

Erosion of genetic diversity of crop plants has for several decades been making it necessary to develop initiatives for protecting these plant resources. One strategy is in-situ conservation of crop plants. The model currently advanced involves maintaining the varieties to be conserved isolated in reserves, protected from entry of other varieties from elsewhere and cultivated according to ancestral farming practices. Researchers from the IRD and the CIMMYT of Mexico (1) used work previously conducted in Mexico on maize varieties, or landraces, to devise a different, dynamic, model judged more compatible with agricultural development and closer to the real conditions in which these plants diversified under the constant action of farmers (2). Insofar as local landraces are still grown as crops on sufficiently large areas of land, the introduction of others from outside, favouring a certain rate of gene flow, would in fact be a source of diversity rather than a factor of genetic erosion.

Mexico, the cradle of maize cultivation, is where this member of the Graminae family, a descendent of a local wild grass, teosinte, was domesticated and phenotypically diversified by human action, at least 6 000 years B.P. An in-situ conservation programme jointly run by CIMMYT, INIFAP (Mexican National Institute of Research in Forestry and Agriculture and Livestock Breeding) and the IRD, conducted in the central valleys of Oaxaca, enabled the research team to characterize the genetic structure of the different populations of local maize landraces and measured the impact of farming practices on this diversity. They focused on two types of diversity: phenotypic (concerning the morphological characters of the plants) and the genetic diversity (observed using genetic markers).

Study of the populations of maize landraces cultivated in six villages of this central region of Mexico has revealed that the morphological and agronomic characters in the field, such as ear size, kernel colour, or flowering period, vary depending on the farmer. At genome scale, genetic markers have shown strong homogeneity between the maize populations within the same village and, more surprisingly, between distant villages. This means that the local varieties possess a common genetic base. The diversity observed in characters of direct pertinence to farmers would consequently be the result of the latter's personal decisions on seed selection, which they make before each crop cycle.

This region's farmers currently grow the maize according to ancestral practices, established over hundreds of years. The fields are sown from one crop cycle to another with the seeds of the ears from the previous harvest. However, from time to time farmers decide to exchange seed batches with other neighbouring farmers, situated more or less geographically distant, in order to run experimental crops from these seeds. Each farmer thus judges the value of these seeds according to their characteristics. Selection criteria are still a highly individual choice and depend on a number of factors. For instance taste, colour, and cooking quality characteristics come into consideration as culinary criteria, leaf characteristics as forage quality criteria.

This investigation has brought the first genetic proof that these cultivation practices, conducted on a small scale (village and region), are a key element in the evolution of maize and its diversity. In-situ conservation of so-called "farm" varieties of crop plants, following the example of maize, could therefore be perceived in terms not of isolation, but of dynamic means of genetic material flow between the different populations of the same region in which the farmers play a predominant role. They thus appear to cultivate maize population diversity. These populations appear as open genetic systems, maintaining the centres of diversity of this major food cereal. The research conducted in this context on the inter-population gene flow should lead to better assessment of the risks from a spread of genetically modified -transgenic-corn varieties that might possibly be introduced into Mexico among traditional local landraces. Such work could therefore help provide some answers for the public debate which for many years has been running on this issue (3).


Marie Guillaume - DIC

(1) This investigation involved jointly research unit UR 141 "Diversité et génomes des plantes cultivées" and the CIMMYT (International Centre for Maize and Wheat Improvement) located in Mexico City.
(2) See FAS N°32, February 1997.
(3) A symposium on this topic is being held by the Commission for Environmental Cooperation of North America, in Oaxaca, Mexico, 11 March 2004. See

An article has also appeared: M. R. Bellon and J. Berthaud, 2004 - Transgenic maize and the evolution of landrace diversity in Mexico: The importance of farmers' behavior, Plant Physiology, 1st March 2004, 134 (3).


Contacter: Julien Berthaud - IRD UR 141 - " Diversité et génomes des plantes cultivées ", 911 avenue Agropolis, BP 64501, 34394 Montpellier cedex 5. Tel.:33-4-67-41-61-65. Email:

Contacts IRD Communication: Marie Guillaume (editor), Tel.:33-1-48-03-76-07, Email: ; Bénédicte Robert (press officer), Tel.: 33-1-48-03-75-19,

References :

G. Pressoir and J. Berthaud, 2004 - Patterns of population structure in maize landraces from the Central Valleys of Oaxaca in Mexico, Heredity, 92, 88 - 94 (01 February 2004)
G. Pressoir and J. Berthaud, 2004 - Population structure and strong divergent selection shape phenotypic diversification in Maize landraces, Heredity, 92, 95-101 (01 February 2004)
M. R. Bellon, J. Berthaud et al., 2003 - Participatory landrace selection for on farm conservation : an example from the Central Valleys of Oaxaca, Mexico, Genetic Resources and Crop Evolution, 50, 401-416

To obtain illustrations concerning this research
Contact Indigo Base, IRD picture library, Claire Lissalde or Danièle Cavanna, Tel.: 33-1-48-03-78-99, Email:

The illustrations can be viewed on :

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.