Gene flow between coexisting oilseed rape varieties and rape volunteers cannot be solely studied on field experiments. Previously. a model (GENESYS: [Colbach, N., Clermont- Dauphin, C., Meynard, J.M., 2001. GeneSys: A model of the inluence of croping system on gene escape from herbicide-tolerant rapeseed crops to rape volunteers. I. Temporal evolution of a population of rapeseed volunteers in a field. Agric. Ecosyst. Environ. 83, 235-253; Colbach, N., Clermont-Dauphin, C., Meynard, J.M., 2001. GeneSys: A model of the inluence of croping system on gene escape from herbicide-tolerant rapeseed crops to rape volunteers. II. Genetic exchanges among volunteer and cropped populations in a small region. Agric. Ecosyst. Environ. 83, 255-270]) was therefore developed to quantify the effects of cropping systems on gene flow from rapeseed varieties to rape volunteers in time and in space. In the present work, the model was improved to simulate simultaneously the spread of three different genes. Four different regions were then surveyed from 1995 to 2001: one region comprised only farmers' fields with frequent rape crops in the rotations; the three remaining regions comprised the French experimental GMO platforms with 2-3 GM and one non-GM rape variety as well as neighbour farmers' fields. For all these fields, cropping history was recorded during the study and for at least three previous years. In the first region, rapeseed volunteers were assessed in crops as well as on road and field marains in 1 year. In the three remaining regions, volunteer densities and genotypes were assessed each year after harvest and rape harvest genotypes determined during 2 years. The comparison of simulated and observed data showed that the model usually satisfactorily predicts volunteer densities and genotype proportions as well as harvest genotype proportions. Several problem areas were identified: volunteers in road and field margins were usually badly predicted; the model frequently underestimated volunteer densities in spring crops; gene flow via pollen dispersal was underestimated when the distance from the pollen source increased. The model was then used to Simulate the spread of herbicide-tolerant rape volunteers in a region grown with one or several GM herbicide-tolerant varieties and more particularly the stacking of several resistance genes in individuals. The simulations also showed the contamination of the harvest of non-GM rape varieties by GM seeds resulting from gene flow. (c) 2004 Elsevier B.V. All rights reserved.
