Field Simulation of Transgenic Seed Admixture Dispersion in Maize with a Blue Kernel Color Marker

Seed purity in maize (Zea mays L.) has been almost neglected in coexistence studies. In this study, the impact of initial seed admixture was assessed in 20 fields (17 integrated pest management [IPM] and three organics) in the Swiss Midlands in 2007 and 2008 and a seed admixture threshold was proposed. The blue-kernelled French hybrid Adonis was used as a color-marker admixture. In 2009, its pollen production was assessed relative to those of two commercial hybrids (LG 32.12 and DKC 2960) that released 41 and 62%, respectively, less pollen than Adonis. Adonis was added to commercial yellow hybrid seeds at a proportion of 1:99. On average, the seed admixture at harvest was 2.8 times greater than the seed admixture at sowing. The greater pollen production of Adonis relative to yellow hybrids explained the increase in the harvest seed admixture via pollen flow. The lesser average kernel number per plant of Adonis (316 kernels plant(-1); 451 kernels plant(-1) for commercial hybrids) buffered the effect of pollen flow. Correcting for these differences provided simulation results compatible with the Hardy-Weinberg equilibrium for a trait with 1% frequency. When compared among the 20 field situations, (i) flower synchrony between admixture and commercial subpopulation and (ii) vertical distance between tassels of Adonis and ears of the commercial hybrid contributed most to field differences. To guarantee the 0.9% threshold imposed by the European Union for non-genetically-modified product, the seed admixture threshold should range between 0.2 and 0.5%.