Effect of family size and number on the accuracy and precision of the estimates of genetic parameters in the IUFRO Douglas-fir provenance-progeny trial

Abstract

Effect of sample size on accuracy and precision of provenance mean height, genetic variance components and heritability were studied in a 10-year-old IUFRO Douglas-fir [ Pseudotsuga menziesii ( (Mirb.) Franco] provenance-progeny trial. Estimates of these parameters, and their respective standard errors, were computed for different numbers of families within provenances, and trees within families. Accuracy of mean height estimates did not differ appreciably with changes in either family number or size. However, precision of the mean height estimates was affected by changes in the family size. Both accuracy and precision of additive genetic variance and heritability estimates were greatly influenced by the number and size of families. Based on estimates obtained empirically, optimum sample sizes for the number of families per provenance and the number of trees per family was proposed and compared with those obtained from analysing the data as a provenance trial (i. e., no family structure). The optimization of sample size will improve the efficiency of provenance-progeny testing.