*4.4. Breeding Aspects*

Large-scale mass selection can be done well by using a mixture with high aggressiveness. In this case, an efficient negative selection should only be done. This should also be measured otherwise, so much aggressiveness can be lost that the necessary high selection pressure cannot be provided.

In an advanced stage of breeding, it is necessary to determine the amount of resistance in the given variety. As responses to DI, FDK and DON may differ, we should test them with more isolates (inocula) with different levels of aggressiveness. This provides a better answer in phenotyping for genetic studies, cultivar registration tests or any other tasks that need high preciosity.

It is not thought that one should be anxious about high aggressiveness to diminish resistance differences. The results prove the contrary; the resistance differences are much higher at high aggressiveness. The papers using high on very high aggressiveness prove this clearly [17,19,22,27,42,58]. Therefore, to identify DON overproduction or DON resistance at least medium or high aggressiveness is needed [69]. A test [15] with four isolates against FHB on 26 genotypes with variable aggressiveness showed that highly resistant lines and cultivars like Sumai 3 and the Szeged line Sgv/NB/MM/Sum3 yielded 0.9 mg/kg DON, the most susceptible line gave between 110 and 433 mg/kg. It seems that in spite of the high aggressiveness the differentiation of the genotypes remained good, even much better than at use of lower aggressiveness. When flowering differences are 2–3 weeks between cultivars, because of the changing environmental conditions their correct comparison is often problematic. By using control genotypes with known resistance levels, the managemen<sup>t</sup> of the resistance tests can be done. However, in a genetic analysis, with the same flowering time differences we have problems, because we should treat the whole population as a unit and therefore data are often not comparable and the QTLs are often artifacts. There are also good examples. In the framework of the USWBSI (United States Wheat and Barley Scab Initiative (www:scabusa.org), multilocation ring tests are made every year to test wheat and Barley genotypes where beside DI, also FDK and DON response, is evaluated [82]. We follow this working way since more than 30 years. This is the only way that can secure the lower DON contamination of the next variety generations. We hope only that scientific research will recognize the need.
