Increase of Complex Pathotypes of Erysiphe graminis f. sp. hordei in Two-Component Mixtures of Spring Barley Cultivars

Pathotypes of Erysiphe graminis f. sp. hordei were monitored at fortnightly intervals in pure stands of cultivars Golf with Laevigatum resistance gene Ml(La)], Steina with Arabische resistance (Mla 12) and Harry with Monte Cristo resistance (Mla9), as well as in two two-component mixtures Golf + Steina (G + S) and Golf + Harry (G + H)] during the course of mildew epidemics in replicated field trials. The number of colonies per main tiller was also recorded. Virulence complexity (i.e. the average number of virulence genes per isolate with respect to the two matching resistance genes of the mixed cultivars) was always higher in mixtures than in pure stands. Linear regression of complexity on the number of elapsed mildew generations (based on actual temperature data) gave significant b-values (slopes) of 0.039, 0.034, 0.022 and 0.018 on Golf in G + S, Steina in G + S, Golf in G + H and Harry in G + H, respectively. In pure stands, there, was no significant complexity increase on any cultivar. The fact that b-values were higher on either component in the G + S than in the G + H mixture indicates stronger selection for the complex, i.e. two-virulence, pathotypes in G + S. Complex pathotypes had a higher relative fitness than simple (one-virulence) pathotypes in both mixtures but not in the pure stands. The absolute frequencies of complex pathotypes as measured by the area under the colony frequency curve (AUCFC) were higher on all cultivars in mixtures than in pure stands, except on Steina.