It has been suggested that plant species partition soil phosphorus (P) resource to reduce competition. This process could contribute to species diversity and structure in plant communities and might therefore be reflected in associations between the spatial distributions of plants and chemical forms of bioavailable soil P. We assessed the influence of P resource partitioning on the spatial distribution of tree species in a 50-ha census plot of a lowland tropical rain forest on Barro Colorado Island, Panama. We determined potentially bioavailable soil P forms via a sequential procedure involving extraction in 2 mM citrate followed by treatment with phosphatase enzymes. The citrate extractable soil P pool was subdivided into reactive (inorganic) P, monoesterase-hydrolysable P, nuclease-hydrolysable P, and non-hydrolysed P. We hypothesized that significant associations between spatial distributions of tree species and low concentrations of soil P fractions provide evidence of P resource partitioning. Associations were tested against null models of random dispersal (α = 0.05).
Results/Conclusions
We found that 21% of tree species were significantly associated with low levels of at least one soil P fraction, although about 11% of these associations might be false rejections of the null hypothesis due to type I error. No tree species was related to reactive P, while similar amounts of species were associated with the three organic P fractions. We interpret these findings as evidence of P resource partitioning at the community scale, affecting a relatively small proportion of tree species on the plot. Interestingly, legumes (Fabaceae) were most frequently associated with soil P fractions. Legumes are known for high root phosphatase activity and our findings suggest that these exoenzymes are synthesized by legumes to acquire soil organic P, consistent with the predictions of the soil P resource partitioning hypothesis.