Pairwise Rank Superiority of Winter Wheat Genotype for Spring Stand

Abstract

Rank statistics can be used as a simple method of summarizing multienvironment data when the assumptions of uniform variances and main effects across environments are violates. A pairwise rank superiority measure is described for analysis of genotypes across environments when usual statistical measures are invalid or too complex for clear data presentation. Eight years of spring stand data for five winter wheat (Triticum aestivum L. emend. Thell.) genotypes were summarized from two locations with intermittent infection by snow mold [Typhula spp. and Microdochium nivale (Ces. ex Beri. & Vogl.) Samuels & Hailer Fusarium nivale Fr Cesati] using ranking of pairs of cultivars to measure relative superiority across environments. For each cultivar pair, a snow mold tolerant cultivar was compared with a susceptible check or another tolerant cultivar. The percent of trials in which a tolerant genotype was superior to the comparison genotype was compared with the null hypothesis of equal performance for each genotype. The snow mold tolerant hard red winter wheat ‘Survivor’ was found to be superior to ‘Manning’, a locally adapted hard red winter wheat, in 78.8% of the trials in which significant stand reduction had occurred. Survivor was significantly superior to Manning for the ability to establish a spring stand based on a t‐test comparison to the null hypothesis of 50% (P < 0.01). Survivor was also superior to the hard red winter wheat cultivars Blizzard and Weston when significant stand reductions occurred, but Survivor was not superior to the soft white winter wheat cultivar Spragne. Measuring the frequency of superior performance within a set of environments with similar stress allows relatively simple cultivar recommendations to be made to growers despite the presence of genotype × environment interactions.