Data analysis
All data (i.e. δ13C, NSCT and its components) were first tested for normality with Kolmogorov–Smirnov tests to assess the within- and between-subject effects in different tissues of pine (i.e. needle, xylem, phloem) and mistletoe (i.e. leaf, shoot).
For the wrapping experiment (Exp. 1), a linear mixed model (tree replicates as random effect) was used for testing the effects of time (sampling time), irrigation treatment (i.e. non-irrigated vs. irrigated), wrapping treatment (i.e. wrapping vs. non-wrapped), and their interactions on the carbon isotopic composition in different mistletoe tissues (i.e. leaf, shoot). The assessment of residuals normality and homoscedasticity were tested before analysis.
For the removal experiment (Exp. 2) (above the girdled branches), the assessment of residuals normality and homoscedasticity were tested before analysis. A linear mixed model (tree replicates as random effect) was used for testing the effects of time (sampling time), irrigation treatment (i.e. non-irrigated vs. irrigated), removal treatments (i.e. needle removal for phloem, xylem and mistletoe tissues; mistletoe removal for pine tissues; and no removal for both pine and mistletoe tissues), and their interactions on δ13C and NSCT concentrations. For each sampling time point, one-way ANOVA and a Tukey-HSD post-hoc test were used to investigate the difference in δ13C and NSCTconcentrations in different pine and mistletoe tissues under different irrigation and removal treatments. R version 4.1.0 was used for all statistical analyses (R Core Team, 2021).