Propensity for seed-mediated gene flow from potato crops and potential consequences for the coexistence of GM and non-GM potato systems

Potato is a critical crop to European growers, both economically and agronomically as a break crop in the standard cereal rotation. As studies investigating the agronomic performance and environmental impact of disease resistant, GM potatoes come to an end across several sites in Europe, past discussions on achieving the effective coexistence of GM and equivalent non-GM crops have too often focussed on the purported risk of excessive pollen-mediated gene flow. Dependent on the crop in question, the impact of seed loss pre- and/or post-harvest presents a greater challenge to securing efficient coexistence practises. To examine this issue for potato, a total of 51 fields that had been commercially cultivated with potatoes were surveyed in two separate cohorts for post-harvest tuber loss and/or volunteer emergence. Across 17 fields studied, the average post-harvest tuber loss was recorded at 141,758 ± 911 tubers ha−1, with volunteer establishment in the following crop ranging from 400 ± 59 ha−1 to 55,698 ± 47 ha−1. In parallel, by surveying a separate cohort of 34 commercial fields an average of 30,789 ± 2658 volunteer ha−1 was recorded in the subsequent cereal crop, with a repeat survey made after an additional year indicating an 87.2% reduction in this mean number of volunteers across the 34 fields (P < 0.001). Of the additional variables studied only location (P < 0.001), herbicide application (P = 0.037) and potato variety used (P = 0.045) significantly influenced volunteer proliferation. Volunteer fecundity was confirmed with up to 3 tubers produced per 1st generation volunteer, with tuber yield from the 2nd generation volunteers reduced significantly (P < 0.001). Assessments of the tuber lots from these 2nd generation volunteers confirmed their ability to sprout post-dormancy, therefore, indicating the potential for 3rd generation volunteers to emerge. Combined, the datasets confirm the potential for significant seed-mediated gene flow from commercial potato systems; indicating that the regulated 0.9% coexistence threshold would in all probability be compromised if GM potatoes were grown in rotations of 1:4 years or less, in the absence of a comprehensive tuber loss and/or volunteer management system