A quantitative framework describing how biomass allocation varies across plant species, communities, and ecosystems is lacking despite its fundamental importance to basic plant biology, global change studies, and evolutionary and ecological theory. Here we derive a model for biomass allocation using an allometric and engineering approach. This model predicts that foliage biomass will scale as the 3/4–power of stem (and root) biomass and as the 2–power of trunk diameter. These scaling relationships are predicted to be invariant with respect to species phyletic affiliation as well as community latitude or elevation (which may be used as crude surrogate measures of community diversity). A world-wide compendium for tree biomass is used to test these predictions in terms of empirically determined scaling exponents. The data in this compendium are shown to comply with the predictions of our model in every critical respect. The implications of our axiomatic tree biomass allocation pattern are discussed in the broader context of ecological and evolutionary theory.

Key words: allometric theory, biomass allocation, biomechanical theory, ecology, trees