NEEDS AND BENEFITS OF EMPIRICAL POWER-TRANSFORMATIONS FOR PRODUCTION AND QUALITY TRAITS IN FOREST TREE BREEDING

Non-normality in the distribution of individual observations of produc tion and quality traits in forest tree breeding may cause inaccurate s election and overestimation of predicted selection gain. The distribut ion of individual observations of traits such as height, diameter, bra nch diameter, branch angle and number of branches per whorl is not alw ays normal. We investigated how the observations were distributed and to what degree it is possible to improve normality, homogeneity of err or variance and additivity by using empirical power transformations. C omputer simulations showed that a seriously skewed distribution impair s selection efficiency and exaggerates selection gain expectations. If the distribution is heavily skewed, transformation might be worthwhil e. It does not seem possible to offer any general advice about which v ariates should be transformed, but in most cases there seems to be no need of any transformation.