Trends and magnitudes of genotype × environment interaction variance components for yield, quality and agronomic traits among coastal short cycle sugarcane breeding populations

Abstract

Genotype × environment interactions (G×E) can reduce the genetic contribution to the phenotype and increase minimum detectable differences, which eventually reduce selection precision. This study was conducted to determine the magnitudes and trends of G×E variance components for sugarcane yield, quality and agronomic traits in the coastal short cycle breeding programmes of South Africa. Data were collected on the plant, first and second ratoon crops of five series of advanced variety trials arranged in a randomised complete block design, planted at Empangeni and Gingindlovu in KwaZulu-Natal, South Africa. Genotype (G) variance was significant (p < 0.05) and its proportion of the total variance was larger than that of components of G×E, indicating higher precision of selecting superior genotypes. The decreasing trend of G variance for yield and agronomic traits indicated reduced genetic diversity on the later populations. Significant (p < 0.05) genotype × location and genotype × crop-year variance in most traits indicated that genotypes adapted to specific locations and genotypes with different ratooning abilities can be identified. Genotype variability for yield and agronomic traits was significantly (p < 0.05) influenced by genotype × location × crop-year, suggesting that ratooning ability of some genotypes was location dependent. The residual variance was highly significant (p < 0.001) and larger than G and G×E variances.

Keywords:

• genetic variability
• genotype by crop-year
• genotype by location
• variance components