Interplot competition between sugarcane families significantly affects estimates of cane yield and family selection.

摘要

Sugarcane family trials are used in the BSES-CSIRO Plant Improvement program to obtain the best estimates of breeding value of parents for cross-pollination, as well as to determine the best families, based on performance, from which individuals should progress for further testing in a breeding program. Cane yield performance in sugarcane clones is known to be affected by interplot competition effects, especially in single-row plots where individuals compete for resources. However, this has not been explored in sugarcane families. It is not known whether competition exists in family trials and, if so, how ignoring competition affects estimates of cane yield of families and, consequently, which families are selected for further trials. Five family trials from the southern BSES selection program were analysed firstly according to current methods used by breeders to estimate yields of families, and then reanalysed after adjusting for interplot competition between sugarcane families. Interplot competition was significant in all five trials, resulting in biased estimates of cane yield. On average, 10% of families would have been overlooked for selection by breeders in the absence of any correction for the effects of competition. We recommend that BSES-CSIRO sugarcane breeders should include the competition model when assessing family performance to improve estimates of cane yield.