Because organic systems present complex environmental stress, plant breeders may either target very focused regions for different varieties, or create heterogeneous populations which can then evolve specific adaptation through on-farm cultivation and selection. This often leads to participatory plant breeding (PPB) strategies which take advantage of the specific knowledge of farmers. Participatory selection requires increased commitment and engagement on the part of the farmers and researchers. Projects may begin as researcher initiatives with farmer participation or farmer initiatives with researcher participation and over time evolve into true collaborations. These projects are difficult to plan in advance because by nature they change to respond to the priorities and interests of the collaborators. Projects need to provide relevant information and analysis in a time-frame that Sustainability 2011, 3 1207 is meaningful for farmers, while remaining scientifically rigorous and innovative. This paper presents two specific studies: the first was a researcher-designed experiment that assessed the potential adaptation of landraces to organic systems through on-farm cultivation and farmer selection. The second is a farmer-led plant breeding project to select bread wheat for organic systems in France. Over the course of these two projects, many discussions among farmers, researchers and farmers associations led to the development of methods that fit the objectives of those involved. This type of project is no longer researcher-led or farmer-led but instead an equal collaboration. Results from the two research projects and the strategy developed for an ongoing collaborative plant breeding project are discussed.

A participatory plant breeding (PPB) program involving the French farmers' association 'Réseau Semences Paysannes' and the French National Agricultural Research Institute (INRA) at Le Moulon was initiated in 2005. In the process of designing the breeding scheme, we evaluated the impact of farmer selection at an early stage (F 2 ) on bread wheat cross progeny populations. The objectives were to characterize the effect of farmer selection, to evaluate the impact of farmer selection on intra-varietal diversity, to provide farmers with relevant information that they can use to improve their selection practices. Early selection was found efficient for some traits and for some of the 35 F 2 -derived F 3 families. For traits of interest such as thousand kernel weight or grain weight per spike, when the response was significant, it was always positive. For most of the traits studied, the among-family genetic variance increased after selection while the average within-family genetic variance decreased. This study provides the first quantitative results for this PPB program and information that will help optimize it in the future.

In maize breeding, the selection of the candidate inbred lines is based on topcross evaluations using a limited number of testers. Then, a subset of single-crosses between these selected lines is evaluated to identify the best hybrid combinations. Genomic selection enables the prediction of all possible single-crosses between candidate lines but raises the question of defining the best training set design. Previous simulation results have shown the potential of using a sparse factorial design instead of tester designs as the training set. To validate this result, a 363 hybrid factorial design was obtained by crossing 90 dent and flint inbred lines from six segregating families. Two tester designs were also obtained by crossing the same inbred lines to two testers of the opposite group. These designs were evaluated for silage in eight environments and used to predict independent performances of a 951 hybrid factorial design. At a same number of hybrids and lines, the factorial design was as efficient as the tester designs, and, for some traits, outperformed them. All available designs were used as both training and validation set to evaluate their efficiency. When the objective was to predict single-crosses between untested lines, we showed an advantage of increasing the number of lines involved in the training set, by (i) allocating each of them to a different tester for the tester design, or (ii) reducing the number of hybrids per line for the factorial design. Our results confirm the potential of sparse factorial designs for genomic hybrid breeding.