cc_byPiepho, Hans-PeterWilliams, EmlynPrus, Maryna2026-01-292026-01-292024https://doi.org/10.1007/s11295-024-01666-6https://hohpublica.uni-hohenheim.de/handle/123456789/18386Progeny trials in tree breeding are often laid out using blocked experimental designs, in which families are randomly assigned to plots and several trees are planted per plot. Such designs are optimized for the assessment of family effects. However, tree breeders are primarily interested in assessing breeding values of individual trees. This paper considers the assessment of heritability at both the family and tree levels. We assess heritability based on pairwise comparisons among individual trees. The approach shows that there is considerable heterogeneity in pairwise heritabilities, primarily due to the differences in both genetic as well as error variances among within- and between-family comparisons. Our results further show that efficient blocking positively affects all types of comparison except those among trees within the same plot.engResidual maximum likelihoodRandomized complete block designResolvable incomplete block designBest linear unbiased estimation (BLUE)Best linear unbiased prediction (BLUP)Average semivarianceSemivariance630Article2025-11-04