Five open pollinated onion cultivars Red Creole (H), Kaharda (I), Koumassa (A), Sokoto local (G) and Ori (E) were crossed in a complete diallel and their progenies were evaluated in replicated trials, at Sokoto and Talata Mafara, Nigeria, during the 2004/2005 and 2005/2006 seasons. The experiments were randomised as complete block design with three replications. 30 mls of 10 -1 cfu of spore suspension of Alternaria porri prepared by serial dilution was poured into the centre of each plot (sunken bed) immediately after irrigation at 2 weeks after transplanting. The AMMI analysis of variance of the populations indicated genotype recorded highly significant (p < 0.001) mean squares of 6086.23 for disease incidence, environment recorded significant (p < 0.05) mean square of 61.56. The genotype, environment and genotype x environment interaction accounted for 99.1 %, 0.1 % and 0.8% of the G + E + GE variation for disease incidence. For cured bulb yield (market weight) genotype and environment recorded highly significant (p < 0.001) mean squares of 55285864.27 and 1272072.30. Genotype and environment explained 99.2 % and 0.3 % of the treatment variation. Crosses I x H and H x I recorded least disease incidence and greater mean cured bulb yield across seasons and locations, and they were stable. Crosses I x A, E x E, E x G, A x E, E x A, G x I and H x E were early maturing.

Keywords: Biplot, Stability, Resistance, Purple Blotch, Disease, Onions