ece3412-sup-0001-FigureS1.tifimage/tif168KFigure S1. Best Linear Unbiased Estimator (BLUE) of handling aggression with its approximate 95% confidence interval, as a function of measurement order. Measurement order reflect whether a nestling was measured first (order 1) in its brood or whether it had to spent additional time in isolation before being measured. Approximate 95% confidence intervals are double the standard error of the BLUE. Measurement order 1 is the model's intercept and the other BLUEs are calculated relative to this value as a factorial fixed effect (see Table 1).
ece3412-sup-0002-TableS1.docxWord document13KTable S1. Variance-covariance matrix and correlation between three offpsring personality traits: handling aggression (AGG), breathing rate (BR), and docility (DOC). Variances in the diagonal, covariances below the diagonal. The correlation (printed in bold) above the diagonal are the same as given in Table 2. REML phenotypic estimates are the sum of its components. Because the nest-of-origin variance for handling aggression was constrained to zero, some of the covariances are not estimable (n.e.).
ece3412-sup-0003-TableS2.docxWord document13KTable S2. Comparison of heritability (h2) estimated using the animal model, where the additive genetic effects are modeled explicitly, compared with heritability of a model only based on reciprocal cross-fostering. Animal model heritability is as given in Table 1. Cross-foster heritability is based on the same data, but assumes that twice the nest-of-origin variance estimates the additive genetic variance. The inflation of using only the cross-fostered data is calculated as cross-foster h2 divided by animal model h2 where a value larger than 1 indicates that ignoring the pedigree structure inflates the estimate of heritability. The difference in heritability estimates are not statistically significant, as judged using a t-test on the estimates, and their difference should be interpreted as illustration only.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.