Wide adaptation: How wide?

Abstract

Breeding programs aim at improving crop production either within given macroenvironments (for example rainfed vs. irrigated) or in a wide range of growing conditions. The merit of either strategy depends on the range of testing environments and on the definition of stress environment. When environments with average yields of 3–4 t/ha are defined as ‘stress environments’, selection for stress conditions can be successfully conducted under optimum conditions. However, when the stress environment has a much lower yield potential (0.5–2.0 t/ha), direct selection in the target environment is the most efficient strategy.

A review of data on ‘widely adapted genotypes’ supports this conclusion.

The relative magnitude of heritability in stressed and non-stressed environments is not sufficient to choose the optimum environment for selection, because phenotypic differences can be of opposite sign in different environments.

The role of constitutive characters as analytical tools in breeding for yield stability in stress environments is briefly discussed.