Stability of resistance to <i>Phytophthora infestans</i> in potato: an international evaluation

Ten institutions in nine countries joined together to test the stability of resistance of 14 potato genotypes to the oomycete pathogen Phytophthora infestans in three separate trials. Seven of the genotypes were tested in one trial involving seven locations, and all 14 were tested in two subsequent trials, each involving eight locations. Stability of resistance was tested with nonparametric tests and with an additive main effects and multiplicative interaction (AMMI) model. Overall, resistance to P. infestans was robust; resistant genotypes were consistently resistant in all locations and trials. The nonparametric analysis indicated that specific genotypes were basically stable across sites for resistance. In trial 3, the Z statistic for overall stability was significant at 0·05%, indicating a significant level of interaction across the trial, but there were no significant interactions for specific genotypes in this trial. The genotype by environment (G × E) effect of the AMMI model was highly significant in both trials, but the mean square of G × E was less than 10% of the genotype effect in each trial. The first two principal components (PCA1 and PCA2) of the AMMI analyses together explained 75 and 80% of the interaction effects in trials 2 and 3, respectively. Based on both nonparametric and AMMI analyses, Ecuador and Argentina were locations of relatively high interaction effects for both trials 2 and 3, although in Ecuador this interaction was not associated with any particular potato genotype. Other locations also had high interaction effects, but these occurred in only one trial. The genotypes Chata Blanca and, to a lesser extent, Torridon were relatively unstable in trials 2 and 3, but in the case of Torridon, resistant, this did not represent a significant loss of resistance.