Stability of barley genotypes for earliness, yield and resistance to leaf scald disease in Ethiopia

Scald, caused by Rhynchosporium secalis (Oud) Davis, is the most widely distributed and destructive disease of barley (Hordeum vulgare) in Ethiopia. Developing genetic resistance in early maturing genotypes is a viable long term strategy to control the disease under unpredictable environments. In 2012 a field experiment was conducted to evaluate thirty-six barley genotypes for barley leaf scald disease resistance, earliness and yield performance in four locations (Mekelle, Korem, Debre-Birhan and Istayish) in the north and central part of Ethiopia. The experimental design was simple lattice with two replications and two spray treatment managements (sprayed and non-sprayed). In addition to the field experiment, data obtained from three environments from the previous experiment were incorporated to obtain deep insight in to the overall performance pattern. Data collected on yield and other agronomic traits including days to heading, days to maturity and disease score were subjected to different statistical analysis tools such as analysis of variance, regression analysis and additive main effect and multiplicative interaction (AMMI) model to evaluate consistency of scald resistance and maturity in barley in diverse environments in order to more effectively breed for stable resistance and earliness. The combined analysis of variance for 14 traits across three treated environments revealed that all but the most important traits central to this study, including days to heading, days to maturity, scald disease and grain yield had shown significant variation for location, genotype and genotype by location interaction. There was no significant variation among genotypes for spray treatments (management) in days to heading and days to maturity except for scald and grain yield. Location effect was the predominant source of variations accounted 70.32%. Besides, the broad-sense heritability estimates of heading days (0.30 on individual plot basis and 0.70 on entry mean across environment) pointed out that earliness is influenced by genotype. So, the possible reasons for most early maturing genotypes to behave susceptible could be early exposure of seedling to the heaviest release of spores from infected residues & adaptation pattern. Sprayed plot yielded up to 0.5 tonne per hectar more than unsprayed. This shows a major effect of scald on yield. The study had also discovered mean performance of genotypes for days to heading across seven environments to be varied between 71 and 96 of which G1, G26 and G34 were found to be early among others. Similarly, for grain yield the variation was between 2.56 and 4.42tha-1 of which G1 had the highest grain yield followed by G30. On contrary, G25 had the lowest grain yield. From trait association analysis, grain yield had significant (P<0.001) positive correlation with biological yield, harvest index as well as seed per spike. Therefore, biological yield and numbers of seeds per spike had considerable contribution for yield to increase and can be used for indirect selection of high yielding genotypes.With regard to scald disease severity%, the lattice adjusted average performance was between 8.83% and 60.03% of which G6 had the lowest score. G4 on the contrary had the highest score. Though it requires further investigation, results suggested that three mega environments have been detected 1. Mekelle, Korem & Ayba, 2. Debre-birhan and 3. Istayish. The existence of high genotype by environment interaction at all environments was identified from the differential response of genotypes rank across locations. As a result, stability of the tested genotypes was determined by AMMI stability value (ASV) and yield stability index (YSI). Hence, based on AMMI2 biplot and genotype mean rank and AMMI stability value (ASV), G1, G6, G7, G15 and G34 were found stably combined scald resistance, earliness, and high yield across all environments. In summary, genotype by environment interaction can be minimized through selection of widely adapted genotypes.