Optimum replications and crop-years for sugarcane genotype trials at Dwangwa Sugar Estate in Malawi.

摘要

The ability to detect significant differences among genotypes in trials depends on the variance associated with the means. Dwangwa Sugar Estate, Malawi, imports genotypes from the South African Sugarcane Research Institute (SASRI) to test for possible commercial release. The objective of this study was to determine the optimum number of replications and crop-years required for adequate discriminating ability for yield and quality parameters. Data for yield (cane and sugar) and quality (estimable recoverable crystal [ERC %] and fibre %) were collected from two trial series. Each trial had eight replications and was harvested in the plant and four ratoon crops. The data were analysed using the MIXED procedure of the Statistical Analysis System (SAS) to estimate the variance components used to calculate broad-sense heritabilities (BSH). The optimum number of replications and crop-years was modeled using BSH. Yield traits produced more significant variance components for genotype by environment interaction (G x E) than quality traits. Quality traits produced larger BSH than yield traits. Four replications and harvesting in the plant and three ratoon crops provided adequate discriminating ability among genotypes. The resources saved by reducing replications could be used to plant more trials on different soil types to exploit or minimise G x E.