Selection of parents for the Sitka spruce breeding population in Britain and the strategy for the next breeding cycle

Abstract

Simple techniques were used to overcome the problems associated with estimating breeding values for a number of traits prior to calculating parental multi‐trait index values for over 800 plus trees within the Sitka spruce General Breeding Population (GBP). Progeny from different trees were planted in half‐sib tests in different years. It took an 11‐year period involving numerous sites of differing quality to complete the test establishment. The estimation of parental breeding value was restricted to across‐sites analysis involving the mean performance of open‐pollinated progeny collected from each plus tree planted over a maximum of three representative sites; in each case the data from each site were weighted by the family heritability for the trait and site. Standard parameter estimates for family heritability and phenotypic and genetic correlations, and standard values of the control common to all experiments, were calculated by averaging estimates over all across‐sites analysis. Ranking of re‐selected parents to form the GBP was based on index values calculated using multi‐trait index selection. Two hundred and forty reselected parents have been allocated to one of four sub‐lines of equal mean index value. Plus trees have been ranked within each sub‐line based on index value. Full‐sib controlled pollinations according to an amended assortative mating procedure are now planned between selections within each sub‐line. The second generation breeding population will be created by selecting the best individuals within the best families.