Figure 2.

Effects of phylogenetic distance of litter mixtures on C/N-ratio changes and microbial biomass. Analyses (a,b) are based on overyielding, i.e. standardized observations for mixtures of litter species by the performance of each species in isolation. Analyses (c,d) are based on raw data, i.e. unstandardized but permitting inclusion of zero-distances into the analysis. Asterisks (*) denote dependent variables are presented as partial residuals from a multiple regression analysis, i.e. accounting for covariables as outlined in Material and methods (partial residuals are dimensionless; see the electronic supplementary material, Appendices 4 and 5 for original data). Phylogenetic distances were based on the genus age-optimized tree. See the electronic supplementary material, Appendices 1 and 2 for analyses based on the sample size-optimized tree. Note that we do not present mass loss here as phylogenetic distance was excluded from the model for the overyielding analyses.