The two main possibilities for decreasing toxin contamination were investigated in this paper. In the breeding section, we found that for resistance evaluation the ratio of Fusarium-damaged kernels is more important as this has a closer correlation with the deoxynivalenol (DON) content than the extent of Fusarium head blight (FHB). A high variability was found among lines from the non-Fusarium programme. A 50% decrease of DON contamination could be achieved by a simple and consequent resistance control. The tests with the variety candidates proved the same; therefore, the resistance screening for variety registration is an effective means to stop the spreading of highly susceptible genotypes. The resistance breeding programme showed an even larger DON decrease related to regular susceptible control varieties. Fungicide treatments were generally only weakly effective. However, it was shown that the testing methodology was poor, and with the optimal coverage spraying as much as 90% reduction of DON on small plot tests can be achieved. A farm-scale technology was also developed where the DON reduction as a mean for 3 years was higher than 70%, more than double the regular praxis data. To stabilize this efficacy, we need the most powerful fungicides, a nearly horizontal spraying of heads (like Turbo FloodJet nozzles) that receive the necessary coverage and so enough active ingredient to protect heads and the spraying technology should be kept rigorously. A combination of resistance and fungicide application can reduce DON contamination levels to below the European Union limit of 1.25 mg kg(-1) for levels which would otherwise be around 8-10 mg kg(-1). We think that this will solve most of the problems.