Methodological problems in field investigations on Fusarium graminearum infection of wheat

Cereal head blight, caused by Fusarium graminearum, is a major disease prob lem. The pathogen is able to spread spores across large distances within cr ops. For plot experiments in the field with artificial inoculation, it has to be considered whether individual treatments can be efficiently separated from each other by isolation strips. In a 2-years field experiment: with w heat performed in south Germany, strips of different width (2, 4, 8 m) were planted with tall-growing winter rape cultivars. They separated wheat: plo ts artificially inoculated with F, graminearum-infected oat grains from non -inoculated wheat test plots. Disease incidence (% infected ears) and sever ity (infected spikelets per infected ear) were investigated on the basis of plot scores and detailed observations of individual ears. The positive cor relation between disease incidence and severity was only moderate. Thus, bo th traits supply supplemental information about the infection level. The as sessment of cereal Fusarium infections can be seriously obstructed by simil ar visual symptoms due to contemporary infections with other diseases. The isolation strips did significantly reduce the infection on test plots. The isolation effect improved slightly with increasing strip width, but infecti on was not completely eliminated even with 8 m wide strips. In conclusion, adequate field experiments need a properly randomized design and a sufficie nt number of replicates.