Why do tree species affect soils? The Warp and Woof of tree-soil interactions

Many ideas have been advanced regarding how trees affect soils. Enough evidence is now available to evaluate the strength of these ideas and to consider interactions between tree species and soils in an evolutionary context. Forest floor mass commonly differs by about 20% for different species growing on the same site; differences of up to 5-fold have been reported. Litterfall mass and N content commonly differ by 20 to 30%, but larger differences are also common (especially with N-fixing species). The net mineralizaton of soil N typically differs by 50% or more among species, indicating very strong feedback possibilities. We evaluate the evolutionary context of tree effects on soils by considering 3 degrees of coupling of trees to soils: tightly woven connections where the fitness of the tree is enhanced by its effect on soils; loosely woven interactions where selection for tree fitness unrelated to soil properties leads to indirect effects on soils (either enhancing or impairing fitness); and frayed interactions where the effects of trees on soil derive from features of the ecosystem that do not involve direct selection for tree fitness. Evidence supports each of these degrees of interaction for at least some cases, and no single context explains all the interactions between trees and soils. Important areas for further work include: next-generation assessments of the effects of trees on soil suitability for the same (and different) species, and the role of soil organisms in developing and modifying the effects of trees on soils.