The rationale for the following research was to analyse the response of sunflower hybrids to different sowing dates and to evaluate hybrid response to critical environmental conditions. The data used are from an experiment conducted in a location-year combination over a period of two years (2007–09) in southern Italy. Eleven hybrids were tested following a randomized complete block design with three replications at each location-year combination. Eight agronomic characters including seed oil

more »

... tent were recorded. Classification and ordination procedures were used to investigate hybrid performance in relation to three different sowing dates. Combined analysis of variance showed that hybrids, location-year combination, sowing date and their interactions were highly significant for all characters. Hybrid performances were classified by cluster analysis into groups that were differentiable in terms of means and stability. The first three components accounted for 74%, 82%, and 87% of the total variation for the first, second and third sowing date respectively. Plotting component one against component two along Euclidean axes should therefore provide a reasonable representation of the spatial arrangements of hybrid performances in the original multi-dimensional space. The applied statistical method gives full information on hybrid performances similarity.