Resistance of wheat to Fusarium head blight caused by Fusarium gramine
arum and F. culmorum was identified in natural epidemics in 1985 and 1
987 as well after artificial inoculations (1983-1988 and 1984-1987). O
ut of 25 genotypes tested, five were identified with no significant di
fference in head blight scores, but differing significantly in yield a
fter artificial inoculation, i.e. tolerance differences were detected
at different resistance levels. Some genotypes that were similar in yi
eld or head blight scores differed in seed infection severity. Genotyp
es with awns were more susceptible to head blight when tested under na
tural epidemic condition in the field; but this trait did not influenc
e head blight severity in artificial inoculations. Dwarf genotypes wer
e more severely infected by head blight than tall genotypes under natu
ral conditions, but genotypes of different plant height classes were s
imilarly susceptible after artificial inoculations. In the early gener
ations of a breeding programme resistance measured by visual evaluatio
n of artificial inoculation is the most important way to screen. If se
lection of dwarf and awned genotypes cannot be avoided, the higher sus
ceptibility caused by awns and dwarfness under natural epidemic condit
ions can be decreased by a higher level of physiological resistance, a
s variability in physiological resistance is available. In later gener
ations, traits like percentage of seed infection or tolerance can be i
dentified by additionally measuring yield reduction. Stability of dise
ase reaction appears to be connected with resistance level, the most r
esistant genotypes are the most stable, and the most susceptible ones
tend to have more unstable reactions in different epidemic conditions.