Previous analyses of olive (Olea europaea L.) crosses have revealed an important degree of pollen contamination. For this reason, the authors tested the paternity of a set of progenies coming from crosses among different cultivars within the olive breeding program of Córdoba, Spain, using four polymorphic microsatellites. They found that the expected pollen sired 141 (83.4%) of the 169 samples analyzed. The contamination rate was either almost total or almost null within each particular cross considered, not being comparable between different crosses. In a second experiment the authors evaluated the influence of several factors on the success of olive crosses, the type of isolation bag, the timing and number of pollinations, and the cross-compatibility of the parents in a multifactorial assay in 2003. They observed no differences in the type of pollination bag used or the number and timing of pollen additions when they analyzed 145 seeds. The main factor affecting the success of the crosses seems to be the intercompatibility among cultivars, because it had a significant influence on the rate of contamination. In the failed crosses, the authors clearly detected the contribution of more than one cultivar to the paternity. The results obtained here indicate that some knowledge of cross-compatibility relationships in olive is required to design crosses in olive breeding programs effectively. To achieve this objective, the progeny parentage analysis could be of great help because of the high level of pollen contamination found among those studied here.