Experimental design studying decomposition of oak litter
For each pair of the focal oaks, we sampled the litter at leaf fall by gently shaking the branches and collecting the leaves before falling on the ground to avoid contamination by soil microbes and arthropods prior to the experimental treatment. Oak litter was air dried in the lab at ambient temperature for at least 2 weeks. For each tree, four samples of each oak litter (about 10 g in equivalent dry weight) were placed in litter bags (25 cm × 15 cm) with 5 mm-mesh size allowing colonization by microbes and soil fauna (Santonja et al . 2017). Moreover, for each tree, five samples were weighted and oven-dried to estimate the water content of the initial litter, i.e. (oven dry weight-air dry weight)/air dry weight. This ratio permitted to estimate the oven-dry weight equivalents of litter samples in the litter bags. Within each pair of trees, litters from the high-phylogenetic isolation tree were exposed below that tree and below the tree of low phylogenetic isolation, and inversely for the low-phylogenetic isolation tree (Fig. 1c). Therefore, a given litter bag was characterized by phylogenetic isolation of the oak from which the litter was sampled and the oak where the litter was incubated, respectively phylogenetic isolation above- and belowground PIA and PIB. Litter bags were posed at approx. 1 m from the trunk of the focal tree, i.e. close enough to avoid impact from other trees. Additional metal-mesh protection was used to avoid physical disturbance by large mammals (wild boars, humans). We harvested the litter bags twice, after 8 months and after 14 months. This gives 8 litter bags for each pair of oaks and a total of 88 litter bags.