Partial resistance to wheat leaf rust in 18 spring wheat cultivars

Summary

Eighteen spring wheat cultivars were tested in microfields and race nurseries for their partial resistance PR to wheat leaf rust under low and high disease pressure respectively. Large differences existed between the 18 cultivars, Skalavatis 56 being the most susceptible and Ponta Grossa 1 being the most resistant cultivar. Of the three epidemic parameters, disease severity (DS) at the time that the susceptible check was severely diseased and area under the transformed disease severity curve (AUTC) and the logistic growth rate (r), AUTC and DS were highly correlated. Both seemed to be reliable estimators of PR but DS should be preferred for economical reasons. The logistic growth rate seemed to be unsuitable as an estimator of partial resistance.

High and low disease pressure gave similar cultivar ranking. PR can be screened and selected equally well in race nurseries with low space, low time and low cost input as in microfields with high space, time and cost input.

Cultivar differences in development rate had a large impact on the cultivar differences for amount of disease and can therefore greatly bias the estimation of cultivar resistance. The resistance of early cultivars tended to be underestimated whereas the resistance of late cultivars tended to be overestimated. The effect of differences in developmental rate was most pronounced in the flag leaf. It is advisable to avoid the assessment of disease levels on the flag leaf only and to incorporate in the tests several susceptible and resistant checks that cover the range of development rates in the material to be selected, because otherwise selection for resistance will tend to select also for lateness.

Regression of the epidemiological parameters on three components of partial resistance revealed that latency period (LP) is an important factor in determining the resistance observed in the field explaining on average 67% of the observed variation. Adding infection frequency (IF) and urediosorus size (US) to the linear model increased the proportion of the observed variation in the field explained by the components to 80%. This result supports the idea that the components of PR inherit independently, at least, in part.