OPTIMUM DISEASE POTENTIAL FOR EVALUATING RESISTANCE TO STENOCARPELLA-MAYDIS EAR ROT IN CORN HYBRIDS

Field trials were carried out at four localities in South Africa to ev aluate 35 corn cultivars for resistance to ear rot caused by Stenocarp ella maydis. Trials at each locality were split to include inoculated plants and plants infected by natural inoculum. Differences in hybrid disease resistance reactions were recorded, but ranking of genotypes o ver trial sites was poorly correlated. Regression analyses (Y = AX(b)) were used to determine the relationship between disease potential of a trial site (X) and observed disease incidence (Y) within a genotype. Disease potential was quantified as the mean disease incidence over a ll hybrids in a trial. Genotypes could be divided into three categorie s: 1) linearly related to disease potential, 2) high susceptibility de spite a low disease potential, and 3) various degrees of resistance de spite increasing disease potentials. This served to explain the absenc e of constant rankings of hybrids and the often conflicting results wh en genotypes screened at different localities were compared for diseas e resistance. Confidence limits fitted to regression lines showed that screening of hybrids within a disease potential of +/-0.6% to +/-50.6 % was acceptable for determining differences between highly resistant and susceptible hybrids, but the range for distinguishing moderately r esistant (intermediate) hybrids from resistant and susceptible hybrids was limited to 4.4 and 26.3% and 1.9 and 40.9%, respectively. Beyond these points, disease reactions converged, suggesting that at very hig h or very low potentials, the use of cultivars in disease control is l imited.