Mixed model ANCOVA for <i>M. polymorpha</i> grown either in the presence or absence of <i>E. medicae</i> in non-serpentine or serpentine destination soils in the greenhouse¹.

1Only the fixed effects or interactions involving presence/absence of rhizobium are shown; random effects reflecting the main plot, split-plot and split-split-plot levels of organization of the factorial experimental design were not of experimental interest. Bonferroni correction for two hypotheses per test yields alpha = 0.025 as a threshold for significance; F-statistics in bold are significant. †, P<0.05; b, P<0.01; c, P<0.001; d, P<0.0001. Abbreviations: P, plant; R, rhizobium; df, degrees of freedom; n, numerator; d, mean denominator; Ln, natural log transform; num, number.