Seed and Plant Improvement and Breeding Department, Ardabil Agricultural and Natural Resources Research Center, Agricultural Research, Education and Extension Organization, Ardabil, Iran.
Rust diseases continue to cause significant losses to wheat production around the world. Among them, yellow rust caused by Puccinia striiformis f. sp. tritici is an important disease that threatens wheat production in most cool environments. Host resistance, especially race- nonspecific resistance, is the most economical way to manage wheat stripe rust disease. In this study, the effectiveness of different types of resistance was compared in field plots at Ardabil Agricultural Research Station (Iran) during 2011-2013. Yield and yield components along with slow rusting parameters including final rust severity (FRS), apparent infection rate (r), relative area under disease progress curve (rAUDPC) and coefficient of infection (CI) were evaluated for 16 wheat cultivars/lines. In all, five wheat cultivars with race-specific resistance, 10 cultivars with different levels of slow rusting resistance and one susceptible cultivar were included in two treatments; with and without fungicide protection under high disease pressure. Results of combined variance analysis showed significant differences between cultivars/lines, also cultivar/line × year at 1% probability level. Wheat cultivars with slow rusting resistance displayed a range of responses indicating phenotypic diversity. Mean thousand kernels weight (TKW) losses of susceptible, race-specific and slow rusting genotypes were 41, 4.4 and 7.6%, respectively. Mean yield losses of susceptible, race-specific and slow rusting genotypes were 65.6, 7.3 and 15.9%, respectively. In this study cultivars having slow rusting resistance with low values of epidemiological parameters were identified. Also genotypes with low yield component losses, despite moderate disease levels, were characterized. Such genotypes can be used in breeding programs to get improved varieties with high levels of resistance and negligible yield losses. Kernels per spike (KPS) data of two experiments were not enough for comparing losses and need supplementary experiments.