The effect of plot type on the estimation of heterosis in bread wheat (T. aestivum)

Abstract

The demonstration of an economically sustainable level of heterosis for grain yield is a key factor which determines the commercial viability of hybrid wheat production. Identification of sufficient hybrid advantage involves testing hundreds, possibly thousands of hybrid combinations each year. The design of an effective hybrid evaluation program depends on the characterization of genotypexplant density interaction effects. These were examined at Narrabri, N.S.W., using six hybrids and their parents and two commercial hybrids planted in four replicate, randomized complete block design experiments using three different plot types, viz. hill plots, space planted plots and multi-row plots. The level of heterosis for grain yield varied depending upon the plot type, with least heterosis being observed in multi-row plots. While the genotypexplot type interaction was significant for grain yield, it was non-significant for yield heterosis. The relative rankings of hybrids did not alter across plot type. Highly significant phenotypic and genotypic correlations between plot types for yield indicated that each plot type provided comparable information on relative genotype performance. The lesser seed requirement and ease of planting points to hill plots being the preferred plot type for preliminary evaluation of hybrid performance. The modulation of heterosis level by plot type suggests that while hill plots will provide an efficient first evaluation, performance would need to be substantiated in multi-row plots.